07-013

Copyright © 2006 by Laura Alfaro, Areendam Chanda, Sebnem Kalemli-Ozcan, and Selin Sayek

Working papers are in draft form. This working paper is distributed for purposes of comment and discussion only. It may not be reproduced without permission of the copyright holder. Copies of working papers are available from the author.

How Does Foreign Direct Investment Promote Economic Growth? Exploring the Effects of Financial Markets on Linkages

Laura Alfaro Areendam Chanda Sebnem Kalemli-Ozcan Selin Sayek

How Does Foreign Direct Investment Promote Economic Growth?

Exploring the Effects of Financial Markets on Linkages∗

Laura AlfaroHarvard Business School

Areendam ChandaLouisiana State University

Sebnem Kalemli-OzcanUniversity of Houston and NBER

Selin SayekBilkent University

August 2006

Abstract

The empirical literature finds mixed evidence on the existence of positive productivity externalitiesin the host country generated by foreign multinational companies. We propose a mechanism thatemphasizes the role of local financial markets in enabling foreign direct investment (FDI) to promotegrowth through backward linkages, shedding light on this empirical ambiguity. In a small openeconomy, final goods production is carried out by foreign and domestic firms, which compete forskilled labor, unskilled labor, and intermediate products. To operate a firm in the intermediategoods sector, entrepreneurs must develop a new variety of intermediate good, a task that requiresupfront capital investments. The more developed the local financial markets, the easier it is forcredit constrained entrepreneurs to start their own firms. The increase in the number of varietiesof intermediate goods leads to positive spillovers to the final goods sector. As a result financialmarkets allow the backward linkages between foreign and domestic firms to turn into FDI spillovers.Our calibration exercises indicate that a) holding the extent of foreign presence constant, financiallywell-developed economies experience growth rates that are almost twice those of economies with poorfinancial markets, b) increases in the share of FDI or the relative productivity of the foreign firmleads to higher additional growth in financially developed economies compared to those observed infinancially under-developed ones, and c) other local conditions such as market structure and humancapital are also important for the effect of FDI on economic growth.

JEL Classification: F23, F36, F43, O40Keywords: FDI spillovers, backward linkages, financial development, economic growth.

∗An earlier version of this paper circulated under the title “FDI Spillovers, Financial Markets and Economic De-velopment.” Corresponding author: Laura Alfaro, Harvard Business School, 263 Morgan Hall, Boston, MA 02163, lal-faro@hbs.edu. Areendam Chanda, Department of Economics, 2107 CEBA, Louisiana State University, Baton Rouge,LA 70803, achanda@lsu.edu. Sebnem Kalemli-Ozcan, Department of Economics, University of Houston, Houston, Texas,77204, Sebnem.Kalemli-Ozcan@mail.uh.edu. Selin Sayek, Department of Economics, Bilkent University, Bilkent Ankara06800 Turkey, sayek@bilkent.edu.tr. We are grateful to Pol Antras, Bruce Blonigen, Santanu Chatterjee, Ron Davis, PeterHowitt, Michael Klein, Katheryn Niles Russ, Pietro Peretto, our discussants, Yan Bai and Ahmed Mobarak, and to partic-ipants at Stanford Institute for Theoretical Economics Summer Workshop, 2006 NEUDC, 2006 Midwest MacroeconomicMeetings, seminar at University of Oregon, 2006 Society of Economic Dynamics Meetings, and 2006 DEGIT Conferenceat Hebrew University, for useful comments and suggestions.

1 Introduction

There is a widespread belief among policymakers that foreign direct investment (FDI) generates positive

productivity effects for host countries. The main mechanisms for these externalities are the adoption

of foreign technology and know-how, which can happen via licensing agreements, imitation, employee

training, and the introduction of new processes, and products by foreign firms; and the creation of

linkages between foreign and domestic firms. These benefits, together with the direct capital financing it

provides, suggest that FDI can play an important role in modernizing a national economy and promoting

economic development. Yet, the empirical evidence on the existence of such positive productivity

externalities is sobering.1

The macro empirical literature finds weak support for an exogenous positive effect of FDI on eco-

nomic growth.2 Findings in this literature indicate that a country’s capacity to take advantage of FDI

externalities might be limited by local conditions, such as the development of the local financial markets

or the educational level of the country, i.e., absorptive capacities. Borensztein, De Gregorio, and Lee

(1998) and Xu (2000) show that FDI brings technology, which translates into higher growth only when

the host country has a minimum threshold of stock of human capital. Alfaro, Chanda, Kalemli-Ozcan

and Sayek (2004), Durham (2004), and Hermes and Lensink (2003) provide evidence that only countries

with well-developed financial markets gain significantly from FDI in terms of their growth rates.

The micro empirical literature finds ambiguous results for the effect of FDI on firm’s productivity.

This literature comes in three waves. Starting with the pioneering work of Caves (1974), the first

generation papers focus on country case studies and industry level cross sectional studies.3 These studies

find a positive correlation between the productivity of a multinational enterprise (MNE) and average

value added per worker of the domestic firms within the same sector.4 Then comes the second generation

1See Blomstrom and Kokko (1998), Gorg and Greenway (2004), Lipsey (2002), Barba Navaretti and Venables (2004),and Alfaro and Rodriguez-Clare (2004) for surveys of spillover channels and empirical findings.

2See Carkovic and Levine (2002).3See also Blomstorm (1986).4A multinational enterprise (MNE) is a firm that owns and controls production facilities or other income-generating

assets in at least two countries. When a foreign investor begins a green-field operation (i.e., constructs new productionfacilities) or acquires control of an existing local firm, that investment is regarded as a direct investment in the balance ofpayments statistics. An investment tends to be classified as direct if a foreign investor holds at least 10 percent of a local

1

studies, which use firm level panel data. However, most of these studies find no effect of foreign presence

or find negative productivity spillover effects from the MNEs to the developing country firms.5 The

positive spillover effects are found only for developed countries.6 Based on these negative results, a

third generation of studies argues that since multinationals would like to prevent information leakage

to potential local competitors, but would benefit from knowledge spillovers to their local suppliers, FDI

spillovers ought to be between different industries. Hence, one must look for vertical (inter-industry)

externalities instead of horizontal (intra-industry) externalities. This means the externalities from FDI

will manifest themselves through forward or backward linkages, i.e., contacts between domestic suppliers

of intermediate inputs and their multinational clients in downstream sectors (backward linkage) or

between foreign suppliers of intermediate inputs and their domestic clients in upstream sectors (forward

linkage).7 Javorcik (2004) and Alfaro and Rodriguez-Clare (2004) find evidence for the existence of

backward linkages between the downstream suppliers and the MNE in Lithuania and in Venezuela,

Chile, and Brazil respectively.8 These results are consistent with FDI spillovers between different

industries.

The purpose of this study is twofold. First, in a theoretical framework, we formalize the mechanism

through which FDI leads to a higher growth rate in the host country via backward linkages, which

is consistent with the micro evidence found by the third-generation studies described above. The

mechanism depends on the extent of the development of the local financial sector. Financial markets

act as a channel for the linkage effect to be realized and create positive spillovers, which is consistent

with the macro literature cited above that shows the importance of absorptive capacities. We are not

firm’s equity. This arbitrary threshold is meant to reflect the notion that large stockholders, even if they do not hold amajority stake, will have a strong say in a company’s decisions and participate in and influence its management. Hence,to create, acquire or expand a foreign subsidiary, MNEs undertake FDI. In this paper, we often refer to the MNE and FDIinterchangeably.

5See Aitken and Harrison (1999).6Haskel, Pereira and Slaughter (2002), for example, find positive spillovers from foreign to local firms in a panel data set

of firms in the U.K.; Gorg and Strobl (2002) find that foreign presence reduces exit and encourages entry by domesticallyowned firms in the high-tech sector in Ireland.

7Hirschman (1958) argues that the linkage effects are realized when one industry may facilitate the development ofanother by easing conditions of production, thereby setting the pace for further rapid industrialization. He also arguesthat in the absence of linkages, foreign investments could have limited or even negative effects in an economy (the so-calledenclave economies).

8See also Kugler (2006).

2

aware of any other study that is consistent with both micro and macro empirical evidence.

In a small open economy, final goods production is carried out by foreign and domestic firms,

which compete for skilled labor, unskilled labor, and intermediate products. To operate a firm in the

intermediate goods sector, entrepreneurs must develop a new variety of intermediate good, a task that

requires upfront capital investments. The more developed the local financial markets, the easier it is

for credit constrained entrepreneurs to start their own firms. The increase in the number of varieties of

intermediate goods leads to positive spillovers to the final goods sector. As a result, financial markets

allow the backward linkages between foreign and domestic firms to turn into FDI spillovers.9 Our model

also implies the existence of horizontal spillovers in the final goods sector since the greater availability

of intermediate inputs not only benefits the foreign firms but also raises the total factor productivity of

the domestic firms in the final goods sector, thus creating a horizontal spillover as an indirect result of

the backward linkage.

In the second half of the paper, we use the model to quantitatively gauge how the response of growth

to FDI varies with the level of development of the financial markets. To the best of our knowledge,

this paper is unique in this respect. We find that a) holding the extent of foreign presence constant,

financially well-developed economies experience growth rates that are almost twice those of economies

with poor financial markets, b) increases in the share of FDI or the relative productivity of the foreign

firm leads to higher additional growth in financially developed economies compared to those observed in

financially under-developed economies. The calibration section, additionally, highlights the importance

of local conditions such as market structure and human capital, the so-called absorptive capacities,

for the effect of FDI on economic growth. For example, we find larger growth effects when goods

produced by domestic firms and MNEs are substitutes rather than complements. By varying the

relative skill endowments—while assuming that MNEs use skilled labor more intensively—we obtain

results consistent with Borensztein, De Gregorio, and Lee (1998) who highlight the critical role of

9In our model, linkages are associated with pecuniary externalities in the production of inputs. In contrast to knowledgespillovers, pecuniary externalities take place through market transactions, see Hirschman (1958). Hobday (1995), in a casestudy of developing East Asia, finds many situations in which MNEs investment created backward linkages effects to localsuppliers.

3

human capital.

Theoretical models of FDI spillovers via backward linkages include Rodriguez-Clare (1996), Markusen

and Venables (1999), and Lin and Saggi (2006). None of these models investigate the critical role played

by local financial markets and neither do they focus on the dynamic effects of FDI spillovers. Instead,

these are static models. Our model closely follows Grossman and Helpman’s (1990, 1991) small open

economy setup of endogenous technological progress resulting from product innovation via increasing

intermediate product diversity. We modify their basic framework to incorporate foreign- owned firms

and financial intermediation. The standard Grossman-Helpman setting is preferred since it provides the

most transparent solution. Further, models of FDI such as the ones mentioned above also use the inter-

mediate product variety structure in a static setting, thus making it a natural choice when moving to a

dynamic framework.10 Recently, Aghion, Howitt, and Mayer-Foulkes (2005) have modeled technology

transfers with imperfect financial markets in a Schumpeterian growth model. Their model is different

than ours in the sense that they focus on credit constraints impeding international technology transfers

(and hence international convergence), while we focus on the role of financial markets easing the credit

constraints and allowing linkages between multinational firms and local suppliers in the host country

to materialize. Thus, we are concerned with linkages within an economy once FDI has taken place. In

a related paper that is closer to the spirit of our paper, Aghion, Comin, and Howitt (2006) develop

a model that highlights the role of local savings in attracting and complementing foreign investment

which spurs innovation and growth.

The importance of well-functioning financial institutions in augmenting technological innovation

and capital accumulation, fostering entrepreneurial activity and hence economic development has been

recognized and extensively discussed in the literature.11 Furthermore, as McKinnon (1973) stated, the

development of capital markets is “necessary and sufficient” to foster the “adoption of best-practice tech-

nologies and learning by doing.” In other words, limited access to credit markets restricts entrepreneurial

development. In this paper, we extend this view and argue that the lack of development of the local

10Gao (2005) also incorporates FDI into a growth model that closely follows Grossman and Helpman (1991). The authorneither models the role of domestic financial markets nor relates the model to empirical evidence.

11See Goldsmith (1969), Greenwood and Jovanovic (1990), and King and Levine (1993a,b), among others.

4

financial markets can limit the economy’s ability to take advantage of potential FDI spillovers in a

theoretical framework, a premise which is already supported by empirical evidence. Our results on

the importance of the financial markets thus contributes to an emerging literature that emphasizes the

importance of the local policies and institutions in limiting the potential benefits that FDI can provide

to the host country.12

The rest of the paper is organized as follows. Section 2 presents the model. Section 3 performs a

calibration exercise using values for the parameters from the empirical literature. Section 4 concludes.

2 The Model

2.1 Households

Consider a small open economy. The economy is populated by a continuum of infinitely lived agents of

total mass 1. Households maximize utility over their consumption of the final good,

Ut =∫ ∞

te−ρ(τ−t) log u(Cτ )dτ, (1)

where u(.) is a continuously differentiable strictly concave utility function, ρ is the time preference

parameter, and Cτ denotes consumption of the final good at time τ . The final good, denoted by Yt,

is a numeraire and is freely traded in world markets at a price pt which we normalize to 1. The total

expenditure on consumption is thus given by Eτ = pτCτ = Cτ . Households maximize utility subject to

the following intertemporal budget constraint,

∫ ∞

te−r(τ−t)Eτdτ ≤

∫ ∞

te−r(τ−t)wτdτ +At, (2)

where At denotes the value of the assets held by the household at time t, and wτ is the wage income.

The intertemporal budget constraint requires that the present value of the expenditures, Eτ , not exceed

the present value of labor income plus the value of asset holdings in the initial period. The solution

12For example, lack of adequate contract and property rights enforcement can limit the interaction between foreign andlocal firms. A foreign firm can decide instead of buying inputs in the host country to produce them within the boundariesof the firm or import them, restricting their local activities to hiring labor. See Antras (2003), and Lin and Saggi (2006).

5

of this standard problem implies that the value of the expenditures must grow at a rate equal to the

difference between the interest rate and the discount rate. However, if this rate of growth of expenditure

is different from the endogenous rate of growth of the economy then either the transversality condition

is violated or the economy no longer remains a small open economy. To rule out these possibilities, we

assume that households are credit constrained and can borrow at most a fixed fraction of their current

income. Further, we assume that this constraint is binding, and therefore the actual rate of growth of

expenditures is proportional to the rate of growth of income:13

.E

E∝Y

Y

2.2 Production

2.2.1 The Final Goods Sector

Final good production combines the production processes of domestic and foreign firms denoted re-

spectively by Yt,d and Yt,f , which are not traded. Let pt,d and pt,f denote their respective prices. The

aggregate production function for this composite final good is given by,

Yt = [Y ρt,d + µY ρ

t,f ]1/ρ, (3)

where ρ ≤ 1 and ε = 1/ (1− ρ) represents the elasticity of substitution between Yt,d and Yt,f . We do

not model the decision of foreign firms to enter the market. Therefore, the aggregator of foreign and

domestic firms’ production serves as an artifact that allows us to capture the interaction of foreign and

domestic firms in an economy.14 We can exogenously vary µ to capture realistic shares of foreign and

domestic firms in the final output. If ε = ∞, foreign and domestic firms produce perfect substitutes;

ε = −∞, they produce complements. If ε = 1, the production function for the final good becomes Cobb

13This is only an assumption of convenience since, as we will see later, the entrepreneurs are also credit-constrainedand we would rather treat both groups the same to rule out any gains from arbitrage. This assumption also ensures thatthe consumption side of the economy has no implications for the production side. Hence, there is no difference betweenassuming a household cannot borrow forever and a household cannot borrow over a certain fraction.

14For a similar setup, see Markusen and Venables (1999).

6

Douglas.

Profit maximization yields,

pt,fpt,d

= µ

[Yt,dYt,f

]1−ρ. (4)

The cost function is given by,

C (Yt, pt,f , pt,d) = Yt

[p1−εt,d + µεp1−ε

t,f

] 11−ε

.

Setting the price equal to marginal cost,

1 =[p1−εt,d + µεp1−ε

t,f

] 11−ε

,

which allows us to derive an expression between the price of the domestic firm and foreign firm goods,

pd =(1− µεp1−ε

f

) 11−ε

. (5)

2.2.2 Foreign and Domestic Firms Production Processes

Both foreign and domestic firms’ production processes combine unskilled labor, skilled labor, and a

composite intermediate good. The intermediate good is assembled from a continuum of horizontally

differentiated goods. Unskilled and skilled labor are not traded and available in fixed quantities L and

H, correspondingly. Competition in the labor market ensures that unskilled and skilled wages, wt,u and

wt,s, are equal to their respective marginal products. To capture the importance of proximity between

suppliers and users of inputs, we assume that all varieties of intermediate goods are non-traded.15

Domestic production is characterized by,

Yt,d = AdLβdt,dH

γdt,dI

λt,d, (6)

15This is a common assumption used to capture transportation costs or local content requirements; see Grossman andHelpman (1990), Markusen and Venables (1999) and Rodriguez-Clare (1996). Alternatively, one could assume that thereare some intermediate goods that are tradable and others that are non-tradable. Our results will hold as long as eachintermediate good enters both domestic and foreign production functions with the same intensity.

7

with 0 < βd < 1, 0 < γd < 1, 0 < λ < 1 and βd + γd + λ = 1. Lt,d, Ht,d, and It,d denote, respectively,

the amount of unskilled labor, skilled labor, and the composite intermediate good used in domestic

production at any instant in time, and Ad represents the time invariant productivity parameter.

Foreigners directly produce in the country rather than license the technology. The industrial orga-

nization literature suggests that firms engage in FDI not because of differences in the cost of capital but

because certain assets are worth more under foreign than local control. If lower cost of capital were the

only advantage a foreign firm had over domestic firms, it would still remain unexplained why a foreign

investor would endure the troubles of operating a firm in a different political, legal, and cultural envi-

ronment instead of simply making a portfolio investment. Graham and Krugman (1991), Kindleberger

(1969), and Lipsey (2003) show that investors often fail to bring all the capital with them when they

take control of a foreign company; instead, they tend to finance an important share of their investment

in the local market. An investor’s decision to acquire a foreign company or build a plant instead of

simply exporting or engaging in other forms of contractual arrangements with foreign firms involves

two interrelated aspects: ownership of an asset and the location to produce.16 First, a firm can possess

some ownership advantage—a firm-specific asset such as a patent, technology, process, or managerial or

organizational know-how—that enables it to outperform local firms. And this is one of the reasons why

researchers fail to find evidence of horizontal spillovers since this means that a foreign firm will seek to

use this special asset to its advantage and prevent leakages of its technology. Hence, we model potential

benefits from FDI as occurring via linkages and not through technology spillovers. Second, domestic

factors, such as opportunities to tap into local resources, access to low-cost inputs or low-wage labor,

or bypass tariffs that protect a market from imported goods can also lead to the decision to invest in a

country rather than serve the foreign market through exports.17

Since our objective in this paper is to understand the effects of foreign production on local output

and the role of financial markets, and not the decision to invest abroad, we model the frictions of doing

16This approach to the theory of the multinational firm is also known as the OLI framework— ownership advantage,localization, internalization. See Dunning (1981).

17For models that endogenize the FDI decision, see Helpman (1984), Markusen (1984), and Helpman, Melitz, and Yeaple(2004).

8

business in the domestic economy with the parameter φ.18 Thus, foreign firms use the following Cobb

Douglas production function,

Yt,f =AfφLβf

t,fHγf

t,fIλt,f , (7)

with 0 < βf < 1, 0 < γf < 1, and βf+γf+λ = 1. Like before, Lt,f ,Ht,f ,and It,f denote, respectively, the

amount of unskilled labor, skilled labor, and the composite intermediate good used in foreign production

at any instant in time, and Af represents the time invariant productivity parameter.

Unskilled and skilled labor have different shares within the domestic and foreign production, though

the total labor share is assumed to be the same across both types of firms. This reflects the common

observation that the share of labor tends to be around two-thirds of total factor payments while at the

same time permitting different skill intensities within domestic and foreign production. A corollary of

assuming the same total labor share is,

γf − γd = βd − βf . (8)

The composite intermediate good is assembled from differentiated intermediate inputs. Following

Ethier (1982), we assume that, for a given aggregate quantity of intermediate inputs used in the final

good production, output is higher when the diversity in the set of inputs used is greater. This spec-

ification captures the productivity gains from increasing degrees of specialization in the production of

final goods.

It,d = It,f = It =[∫ n

0xαt,idi

]1/α

, (9)

where xt,i is the amount of each intermediate good i used in the production of the final good at time t,

and n is the number of varieties available. Let pi denote the price of a variety i of the intermediate good

x. The CES specification imposes constant and equal elasticity of substitution (1/(1−α)) between a pair

18Burnstein and Monge-Naranjo (2005) assume taxes on foreign firms to be the barrier in each country. We allowa broader interpretation, as foreign firms need to bear a wide range of costs/risks of doing business abroad, includingsovereign risk, taxes, and infrastructure and dealing with different institutions and cultures. We also considered analternative scenario where MNEs receive a net price pf/φ where φ > 1, reflecting these disadvantages, obtaining similarresults.

9

of goods. Each variety of intermediate good enters the production function identically and the marginal

product of each variety is infinite when xt,i = 0. This implies that the firm will use all the intermediate

goods in the same quantity, thus xt,i = xt. Let Xt = ntxt be the total input of intermediate goods

employed in the production of the final good at time t, then we can rewrite It = n1−α

αt Xt. Domestic

production is given by,19

Yd = AdLβdd H

γdd X

λdn

λ(1−α)α , (10)

and foreign production by,

Yf =AfφLβf

f Hγf

d Xλf n

λ(1−α)α . (11)

Thus, raising the varieties of intermediate inputs n, holding the quantity of intermediate goods con-

stant, raises output productivity. Using the cost function and the fact that in a symmetric equilibrium

all intermediate goods are priced similarly, pi = px, we can write the equilibrium conditions for the

domestic and foreign firms respectively as,

pd =A−1d β−βd

d γ−γdd

λλwβdu w

γds p

λxn

λ(α−1)α , (12)

pf =φA−1

f β−βf

f γ−γf

f

λλwβfu w

γfs p

λxn

λ(α−1)α . (13)

2.2.3 The Intermediate Goods Sector

The intermediate goods sector is characterized by monopolistic competition. There exists an infinite

number of potential varieties of intermediate goods, but only a subset of varieties is produced at any point

in time as entrepreneurs are required to develop a new variety. Since the set of potential intermediate

goods is unbounded, an entrepreneur will never choose to develop an already existing variety. Therefore,

variety i of x is produced by a single firm which then chooses the price pi to maximize profits. Firms

19Since we will focus exclusively on the balanced growth path, we omit the time subscript for the rest of the paper.

10

take as given the price of competing intermediate inputs, the price of the final good, and the price of the

factors of production. In a symmetric equilibrium all intermediate goods are priced similarly, pi = px.

Hence, profit maximization in every time period for each supplier of variety i implies,

max πi = pxxi − cx(wu, ws, xi)xi, (14)

where cx(wu, ws, xi) represents the cost function and xi = xd + xf is the sum of the demand for the

intermediate product i by domestic and foreign firms respectively.

Production of intermediate goods requires both skilled and unskilled labor according to the following

specification,

xi = LδxiH1−δxi

. (15)

Hence, the cost function for the monopolist is given by,

c (wu, ws, xi) = δ−δ(1− δ)−(1−δ)wδuw(1−δ)s xi. (16)

Profit maximization yields the result that each variety is priced at a constant markup (1/α) over

the marginal cost.20 Hence, the price of each intermediate good is given by,

px = δ−δ(1− δ)−(1−δ)wδuw

(1−δ)s

α. (17)

The fraction that domestic firms spend on all intermediate goods is given by the corresponding share

in the production function, λpdYd. This implies that for each intermediate good, the amount spent by

domestic firms is given by λpdYd/n. Similarly, the amount that foreign firms spend on these goods is

given by λpfYf/n. The sum of amounts spent by foreign and domestic firms is the total revenue of the

intermediate producer,

20See Helpman and Krugman (1985), chapter 6.

11

pxxi =λpdYdn

+λpfYfn

. (18)

Therefore, we can write the operating profits per firm as,

πi =(1− α)n

[λpdYd + λpfYf ] . (19)

What is the value of introducing new intermediate goods and thus the value of the monopolistic

firm? Let vt denote the present discounted value of an infinite stream of profits for a firm that supplies

intermediate goods at time t,

vt =∫ ∞

te−r(s−t)πsds.

Equity holders of the firm are entitled to the stream of future profits of the firm. They make an

instantaneous return of (πt +.v), (profits and capital gain). They can also invest the same amount in a

risk-free bond and receive return rvt (the prevailing market interest rate). Arbitrage in capital markets

ensures that,

π + v = rv ⇒ π +.v

v= r. (20)

Thus, the rate of return of holding ownership shares is equal to the interest rate.21

2.2.4 Introduction of New Varieties and Financial Markets

In order to operate a firm in the intermediate good sector, entrepreneurs must develop a new variety

of intermediate goods. The introduction of each new variety requires some initial capital investment

according to the following specification,

n =K

anθ. (21)

21Note that the arbitrage condition does not contradict our assumption of credit-constrained households since they canchoose to lend to firms or invest in a risk free bond.

12

In contrast to Grossman and Helpman (1991), who assume that new varieties are developed with two

inputs, labor and general knowledge, we opt for one input only, capital, to simplify the analysis.22 Our

main results do not depend on this simplifying assumption. The main implication of this simplification

is that our results are less dependent on the production parameters of the innovation sector. This has

important advantages for our calibration exercise, as the stylized facts of the innovation and imitation

processes are not well documented. Our central argument is that entrepreneurs face difficulties in

obtaining, for example, loans to set up firms and this prevents the creation of backward linkages even

under the presence of FDI. Assuming only capital is used for these setup costs then allows us to focus

better on this issue.

The introduction of a new variety depends on the existing stock of varieties. We introduce the

parameter θ since this allows a more general production structure. A value of θ < 0 suggests a “fishing

out” effect (increasing complexity in introducing new varieties) while a value of θ > 0 implies positive

externalities (“standing on the shoulder of giants”). At this stage, we do not postulate an exact value

of θ.23 This will be pinned down by conditions required to satisfy balanced growth. Finally, a can be

viewed as the level of efficiency in the innovation sector.

The initial capital investment must be financed by borrowing from domestic financial institutions.

The domestic financial system intermediates resources at an additional cost, as in Edwards and Vegh

(1997). This cost reflects the level of development of the domestic financial markets where lower levels

of development are associated with higher costs. These costs manifest themselves in a higher borrowing

rate, i which is greater than the lending rate, r. As King and Levine (1993a) mention, this wedge could

reflect taxes, interest ceilings, required reserve policies, high intermediation costs due to labor regulation,

or high administration costs, etc. This simplification allows us to focus on the main theme of the paper:

the role of financial markets in allowing FDI benefits to materialize. Thus, this assumption should

22Grossman and Helpman (1991) assume that the greater the stock of general knowledge among the scientific community,the smaller the input of human capital needed to invent a new product. They assume

.n = KL/a, where K represents

the stock of general knowledge capital and not physical capital like in our model. Hence, in our model, in incurring setupcosts, intermediate firms do not compete for labor inputs against the final good sector.

23For more on the implications of these alternative assumptions, see Jones (1995). While we do not postulate an exactvalue, we will work under the assumption that θ 6= 1. Also for convenience, the discussion will treat θ as positive.

13

be regarded as a shortcut to more complex modelling of the financial sector. The reader is referred

to the appendix for a cost verification approach following King and Levine (1993b) that yields similar

implications. Our qualitative results will be the same under this framework.

Given nθ and a, if an entrepreneur wants to introduce one variety at any instant in time, the amount

of capital needed will be K = a/nθ, so that n = 1. Therefore, the cost of introducing a new variety is

ia

nθ. (22)

There is free entry into the innovation sector. Entrepreneurs will have an incentive to enter if

ianθ < v. However, this condition implies that the demand for capital will be infinite, which cannot be

an equilibrium solution. Hence, we can rule out this condition ex-ante. If, on the other hand, ianθ > v,

entrepreneurs will have no incentive to engage in innovation. This possibility cannot be ruled out ex-

ante but would lead to zero growth. Therefore, in equilibrium, if there is growth in the number of

varieties it must be the case that,

ia

nθ= v iff n > 0. (23)

This also implies,

v

v= −θ n

n,

i.e., more innovation reduces the value of each firm. Using this expression and the arbitrage condition

in the capital markets, πv + v

v = r, we can rewrite equation (20) as,

π

v− θ

n

n= r. (24)

Using firm profit equation (19), equation (23), and equation (24) we obtain,

(1− α)λia

[pdYdn1−θ +

pfYfn1−θ

]− θ

n

n= r. (25)

In order to simplify, we define Yd

n1−θ = Yd and Yf

n1−θ = Yf as efficiency units of outputs and get,

14

n

n=

(1− α)λθia

[pdYd + pf Yf

]− r

θ(26)

As it is standard in this class of models, the growth rate of varieties, n/n, ultimately pins down the

growth rate of both domestic output and foreign output and thus aggregate output as well.

2.3 General Equilibrium and the Balanced Growth Path

Using the efficiency unit adjusted output levels, Yd and Yf , we can also rewrite equation (4) as,

pfpd

= µ

[Yd

Yf

]1−ρ

. (27)

And we can rewrite equation (21) as,

n

n=

1a

K

n1−θ =1aK,

where K is the capital stock per efficiency unit.

Substituting the price for intermediate inputs (17) into the equilibrium conditions for the domestic

(12) and foreign production (13), and defining efficiency wages for both skilled and unskilled labor

as ws = ws/(n

(1−α)λα

)and wu = wu/

(n

(1−α)λα

), for the domestic and foreign sector, respectively, we

obtain the following expressions,

pd =A−1d β−βd

d γ−γdd

λλ

(δ−δ(1− δ)−(1−δ)

α

)λwβd+δλu wγd+(1−δ)λ

s (28)

pf =φA−1

f β−βf

f γ−γf

f

λλ

(δ−δ(1− δ)−(1−δ)

α

)λ

wβf+δλu w

γf+(1−δ)λs (29)

Equilibrium conditions in the labor market imply that the labor employed by the domestic, the

foreign, and the intermediate goods production processes add up to the total labor supply in the

economy. This implies, for skilled and unskilled labor, respectively,

15

Ld + Lf + nLx = L, (30)

Hd +Hf + nHx = H. (31)

Using the cost functions for the domestic, foreign and intermediate goods sector and Shephard’s

Lemma, we can rewrite these two constraints as,

(βd + δαλ) pdYdwu

+(βf + δαλ) pfYf

wu= L,

(γd + (1− δ)αλ) pdYdws

+(γf + (1− δ)αλ) pfYf

ws= H.

Using our output and wage equations in terms of efficiency units, we can rewrite both constraints

as,

(βd + δαλ) pdYdn1−θ

wun(1−α)λ

α

+(βf + δαλ) pf Yfn1−θ

wun(1−α)λ

α

= L,

(γd + (1− δ)αλ) pdYdn1−θ

wsn(1−α)λ

α

+(γf + (1− δ)αλ) pf Yfn1−θ

wsn(1−α)λ

α

= H.

Along the balanced growth path, labor shares must be constant. This means that(piYin

1−θ)/wjn

(1−α)λα

(i = d, f ; j = u, s) should be constant. There are two ways to achieve this. The first is to assume that

prices, pi, grow at the rate(

(1−α)λα − (1− θ)

)nn . This assumption still implies that pd/pf would be

constant since none of these parameters reflect sectoral differences and the relative price would thus be

driven by other factors (mainly ρ and µ). An alternative would be to impose (1−α)λα = 1− θ. Although

this assumption has the disadvantage of being a knife-edge condition, on the other hand, it does offer

a big advantage. It allows us to back out the value of θ, for which empirical measures are not available

(measures for λ and α on the other hand are easily available from the literature). This condition still

implies that pd/pf would be constant. Therefore, we assume that this condition holds. Thus, we can

rewrite the above as,

16

(βd + δαλ) pdYdwu

+(βf + δαλ) pf Yf

wu= L, (32)

(γd + (1− δ)αλ) pdYdws

+(γf + (1− δ)αλ) pf Yf

ws= H. (33)

Now we can solve for all the endogenous variables and derive the equilibrium balanced growth. In

order to be able to solve the equilibrium growth rate of varieties, nn , we need to solve the set of prices

{pd, pf , wu, ws} and the outputs of the domestic and foreign sectors,{Yd, Yf

}. To solve for the prices

and the outputs, we use equations (4), (5), (28), (29), (32) and (33). These equations can be solved

in a sequential order. The details are presented in the appendix. While we can solve for the FOCs

and derive implicit relationships, because of equation (5), we cannot derive explicit solutions for the

endogenous variables in terms of the parameters.

Our model exhibits some of the standard properties of product variety-based endogenous growth.

Combining the above with equation (26),

n

n=

(1− α)λθia

[pdYd + pf Yf

]− r

θ

we can see that, the scale effect is very much present—larger endowments (L and H) will lead to larger

output and thus to higher growth rate. Furthermore, higher λ, which is the share of intermediate

input costs in final output production, also drives up the growth rate of n by the same reasoning.

Similarly, lower substitutability among intermediate goods (α) increases the growth rate since it raises

the profitability of new intermediate goods. An increase in either, Af , or µ, will lead to a reallocation

of resources away from the domestic firm to the foreign firm. Therefore, the instantaneous effect will

be a decline in domestic firms’ share in output. In the long run, both domestic and foreign firms

will benefit from the higher growth rate. However, in the short-run, the horizontal spillovers in the

final goods sector, which indirectly result from the backward linkages between the foreign firm and the

intermediate goods sector, exist only for the surviving domestic firms. This is an additional contribution

of our setup, which can shed light on why empirical studies fail to find evidence of positive horizontal

17

spillovers for developing countries and even find negative spillovers in some cases.24

Moving on to the role of the financial markets, one can see that the lending rate and the borrowing

rate have negative effects on the growth rate. The negative effect of the lending rate r is standard—a

higher r reflects a greater opportunity cost of investing in a new variety and thus reduces the growth

of varieties. The negative effect of the higher borrowing rate, i, is more novel. It reflects the higher

per unit cost of initial capital investment (because of inefficiencies in financial markets) and thus also

unambiguously reduces the growth rate.

If we restrict ourselves to the special case of where the aggregator is a Cobb Douglas function (or

perfect substitute case), we can solve explicitly for all the endogenous variables. We turn to these next

to get a sense of the qualitative implication of the model.

The Special Case of Cobb Douglas Production Function

The Cobb Douglas case is CES with ρ → 0. We can rewrite the aggregator for the domestic and

foreign output as,

Y = Y1

1+µ

d Yµ

1+µ

f , (34)

Y = Y ηd Y

1−ηf , (35)

where µ = (1− η) /η for simpler notation. Note that profit maximization here implies that

pdpf

=η

1− η

YfYd. (36)

24While our model is very much in the spirit of traditional endogenous growth models, more recently there has been atrend to move towards models that suggest a long run exogenous growth rate with endogenous growth in transition (e.g.see Aghion et al (2005)). This class of models implies, that in the long run, differences in financial markets or extentof FDI would be reflected in transition paths or differences in relative income levels instead. We have not adopted thisscheme a) because focusing on relative income levels abstracts from some of the dynamic spillovers that are interestingwhen one talks about FDI and growth, b) the product variety models do not easily lend themselves to endogenous growthin transition and exogenous growth in the steady state.

18

As for the cost function and equilibrium conditions,

C(y, pd, pf ) = η−η (1− η)−(1−η) pηdp(1−η)f Y = 1,

⇒ pd = η (1− η)(1−η)

η p−(1−η)

η

f . (37)

Recalling the arbitrage condition,

(1− α)λia

[pdYd + pf Yf

]− θ

n

n= r.

Using (36), the previous expression as,

⇒ (1− α)λia (1− η)

pf Yf − θn

n= r. (38)

We can solve the model completely, using equations (28), (29), (32), (33), (36), and (37). The details

are worked out in the appendix. The main equation of interest is (68) in the appendix, renumbered

here as (39)

((1− α)λ

ia

)(wsH

η (γd + (1− δ)αλ) + (1− η) (γf + (1− δ)αλ)

)− θ

n

n= r, (39)

where wsH = ∆ΥΛηdΛ1−ηf (ΦL)Ψβ HΨγ ; Υ = ηη (1− η)(1−η) ; Λd =

(Adβ

βdd γ

γdd

); Λf =

(Afβ

βf

f γγf

f

)/φ;

Φ =η(γd+(1−δ)αλ)+(1−η)(γf+(1−δ)αλ)

η(βd+δαλ)+(1−η)(βf+δαλ) ; Ψβ = η (βd + δλ) + (1− η) (βf + δλ) ; Ψγ = η (γd + (1− δ)λ) +

(1− η) (γf + (1− δ)λ) .

There are a couple of conclusions one can draw from the Cobb Douglas case. First, higher pro-

ductivity of either domestic or foreign firms raises the growth rate in the economy since, all else being

equal, a higher Af and a higher Ad both tend to raise growth. The equation for wsH above, is also

increasing in both L and H. Essentially, an increase in any of these four exogenous parameters increases

19

the profits from introducing new intermediate goods and thus causes the growth rate to rise. However,

note that even if Af increases relative to Ad (i.e. a higher productivity gap between the domestic and

the foreign producer) this will not increase the share of the foreign producer in the market. This is the

clear drawback of using a Cobb Douglas specification. Second, the effect of a higher share of foreign

production, (1− η) , on aggregate growth is ambiguous. The ambiguity can be attributed to the term,

Υ = ηη (1− η)(1−η) which is a U-shaped function of η that is minimized at η = 0.5. For most of the

countries in the world η > 0.5 and for most developing economies it would be near to 1. Therefore,

even if there is a productivity gap between the domestic and the foreign producer, the term Υ, which

is independent of this gap, could drive down the growth rates.

The Special Case of Perfect Substitutes

CES aggregators allow for finite elasticities of substitution. But what if the two outputs were perfect

substitutes? Perfect substitutes is of course a special case of the CES with ρ → 1 (and for simplicity

assume that µ = 1 as well). In this situation, it is easier to bypass the aggregator (since both products

will have the same price and are indistinguishable) and assume that they are traded in the international

market with the world price normalized to 1. One might wonder if the domestic sector would survive at

all given the technological superiority of the foreign firms. However, note that the production function

parameters for the domestic and foreign firms are different allowing for both firms to co-exist while

setting the marginal costs equal to the price of the final good, since the relative marginal costs are not

completely driven by productivity differences. The explicit solution for the perfect substitute case is

worked out in the appendix.

Next, we turn to the calibration exercises, where by using empirical estimates of our parameters we

quantitatively study the comparative static effects we have discussed so far.

20

3 Calibration Exercise

The purpose of the calibration exercise is to study the quantitative growth effects of FDI, focusing on

different levels of financial market development. We begin with a description of the parameters used in

the analysis.

Financial Development: We group countries based on their financial market development levels.

Different measures have been used in the literature to proxy for financial market development. The

broader financial market development measures, such as the monetary-aggregates as a share of GDP

and the private sector credit extended by financial institutions as a share of GDP, capture the extent

of financial intermediation; interest rate spreads, on the other hand, capture the cost of intermediation.

Given that the spread between the lending and borrowing rates better captures the spirit of our model,

we prefer it as the measure for the development of the financial markets.25 We find that the alternative

measures of financial market development, such as the size of the financial market, the share of private

sector credit in total banking activity, and the overhead costs are all highly correlated with interest

rate spreads. Hence, different measures yield similar results.26 The average spread for the low finan-

cially developed (poor) countries, medium financially developed (middle income) countries and the high

financially developed (rich) countries between 2000 and 2003 are 14.5%, 8.5%, and 4.5%, respectively.

Elasticity of Substitution: In our model, ρ relates to the elasticity of substitution between goods

produced by foreign and domestic firms. Evidence regarding the appropriate choice of the elasticity of

substitution parameter ρ is sparse, given that such depiction of final goods production is an artifact

to capture the interaction between foreign and domestic firms. The evidence that is closest to the

spirit of our model is from the consumption literature that uses a constant elasticity of substitution

utility function between varieties of domestic and foreign goods, or between tradable and nontradable

goods. Ruhl (2005) provides a detailed overview of the Armington elasticity, i.e., the elasticity of

substitution between foreign and home goods, and finds that an appropriate value for ρ is around

25Erosa (2001) defines the financial intermediation cost as the resources used per unit of value that is intermediated,which is the total value of financial assets owned by the financial institutions. He measures the financial intermediationcost as the spread between the lending and borrowing rates.

26Alternative measures are from Levine et al. (2000).

21

0.2.27 While our benchmark analysis is based on the CES production function with ρ = 0.2, we also

undertake robustness analysis by allowing ρ to vary between −0.9 and 0.9. In section 3.2, we present

the quantitative characteristics of the model for the Cobb Douglas case (ρ = 1).

The share of intermediate goods in the production of the final good (λ) is assumed to be the same

across the two production technologies. The formulation of the production technology allows setting

the share of the intermediate goods equal to the share of physical capital in final goods production.

Following Gollin (2002), we set this share equal to 1/3. The remaining 2/3 is accounted by skilled

and unskilled labor. The remaining parameters used in the benchmark analysis are chosen such that

those for the domestic firm capture the characteristics of the production technologies available in the

developing countries; whereas, those for the foreign firm capture the characteristics of the production

technologies available in the industrial countries.

Domestic Firms: According to Weil (2004), the share of wages paid to skilled labor is 49% for the

developing countries. We take this value to be that of domestic firms, suggesting that of labor’s 2/3rd

share in final goods production, 49% is due to skilled labor. Therefore, we set the share of skilled labor

in domestic firms, γd, at 32%. In parallel, the share of unskilled labor in domestic firms, βd, is set at

35%. For the benchmark analysis, we set the total factor productivity Ad equal to 1.

Foreign Firms: The share of skilled and unskilled labor costs in output of the foreign firm is

calculated in a similar fashion. Following Weil (2004), the share of wages paid to skilled labor is taken

as 65% in industrial countries. Accordingly, the share of skilled labor in the foreign firm’s production,

γf , is set equal to 40%. Similarly, the share of unskilled labor, βf , is set equal to 27%. Thus, γf > γd.28

As a benchmark, the productivity of the foreign firm, Af , is initially set to be twice that of the domestic

firm following Hall and Jones (1999), who show the productivity parameter for a very large sample of

non-industrial countries is around 45% of the productivity parameter of the U.S. With respect to the

27A wide range of estimates are available from trade and business cycles literatures ranging between 0 and 0.5. Ruhl(2005) argues that a model with temporary and permanent trade shocks can replicate both the low elasticity of substitutionfigures used by the international real business cycle studies and the high elasticity of substitution values found by theempirical trade studies. Such an encompassing model justifies a value of ρ around 0.2.

28As Barba Navaretti and Venables (2004) note, there is ample evidence that foreign firms employ more skilled personnelthan domestic firms. They also tend to be larger, more efficient, and pay higher wages.

22

cost of doing business that the foreign firms face, φ, our benchmark case is one where there is no such

cost. However, note that an increase in the cost of doing business is equivalent to lower productivity of

foreign firms. Thus, by considering variations in relative productivities, we can also infer implications

for the variations in cost of doing business.

Share of Foreign Production: The share of foreign production to total output is not exogenous in

the CES production function case and the choice of µ implicitly determines this share. As such, the

benchmark value for µ is determined to allow for the matching of the relative output values to the real

data. Lipsey (2002) estimates that in 1995 the share of world production due to FDI flows was at best

8%.29 Keeping this in mind, we set µ = 0.1 as our benchmark value since, as we shall see later, it

produces a share of approximately 6%. In the Cobb Douglas production function case, we round off the

share of foreign firms in total output to 5% (i.e. η = 0.95).

Intermediate Goods Sector: Based on the work of Basu (1996), the mark-up is assumed to be 10%,

and hence the value of the reciprocal of (1+mark-up) is given by α = 0.91. Given the lack of any

estimate, the share of unskilled labor in the production of the intermediate goods, δ, is taken as 0.5.

The Stock of Skilled and Unskilled Labor: H and L, respectively, are set following Duffy, Papa-

georgiou, and Perez-Sebastian (2004). The authors argue that there is an aggregation bias caused by

differences in terms of efficiency units of the different types of labor. To overcome this bias, they weigh

the length of education by the returns to schooling, and compute what they call “weighted” labor stock

data. We calculate averages of their data for the low financially developed (poor) countries, medium

financially developed (middle income) countries, and the high financially developed (rich) countries.

Accordingly, we set the ratio of unskilled labor to skilled labor equal to 12 for the poor countries, 9 for

the middle income countries, and 5 for the rich countries. To rule out the possibility of scale effects

driving differences in growth rates, we assume that H + L = 1. The shares of the two factors are

allocated according to these three ratios so that they sum to 1 (e.g., for poor countries H = 0.077 and

L = 0.923).

29Mataloni (2005) finds that foreign owned companies were responsible for 12% of GDP in Australia, 5% in Italy, 7% inFinland, 19% in Hungary, and 22% in the Czech Republic.

23

Additional Parameters: The cost of introducing a new variety, a, is taken to be a free parameter.

The model is calibrated to allow for the financially well-developed country growth rates to match the

U.S. growth rate. Given the fact that the U.S. is often considered to be the technological leader, one

can assume that the productivity of foreign firms in the U.S. is no different than the productivity of

the domestic U.S. firms, so that Af/ (φAd) = 1, to back out a. The U.S. growth rate of real GDP was

approximately 3.5% for the period 1930–2000. This condition and the other parameters above pin down

a = 15 for the CES production function case with ρ = 0.2, and a = 60 for the Cobb Douglas production

function case. We use the value of a = 15 also in the sensitivity analysis of the CES case when we allow

the ρ value to range between −0.9 and 0.9.

The risk-free interest rate is assumed to be 5%. Finally, the parameter capturing the ease of devel-

oping new variety of products, θ, is limited by other parameter choices given the following formulation:

θ = 1− (λ(1− α)/α). Table 1 summarizes all the parameter values.

We consider two scenarios that reflect the benefits of FDI. The first scenario is an exogenous increase

in the share of FDI due to increases in µ. Increases in µ in the CES aggregator lead to a higher share

of foreign output in GDP. This exercise answers the straightforward question: What happens to the

overall growth rate of the economy if the more productive MNE’s produce a higher share of output?

The second scenario is where advances in innovation in the parent country are transmitted through

FDI to the host country. These technological benefits of FDI are captured through the productivity

parameter of the foreign firm (i.e, an increase in Af ). Our initial tests are based on the effects of a

15% increase in the productivity of the foreign firm relative to the domestic firm. Starting with our

benchmark value of Af/φAd = 2, this would mean a new value of Af/φAd = 2.3 (φ = 1 in both cases).

Later on when undertaking sensitivity tests, we consider a range of values between 1.15 and 2.6. The

lower bound of 1.15 is based on Aitken and Harrison’s (1999) finding that as a plant goes from being

domestically owned to fully foreign owned, its productivity increases by about 10% to 16%. Given there

is no consensus in the empirical estimates, in our analysis, we use a wide of values thus providing a

more comprehensive picture.

Returning to the initial assumptions of Af = 2Ad, and φ = 1, note that our consideration of a

24

range for Af/φAd from 1.15 to 2.6 can also be implicitly used to understand the effects of variations

in the cost of doing business, φ, when Af = 2Ad. Thus, Af/φAd = 2.6 would correspond to φ = 0.77,

and Af/φAd = 1.15 to φ = 1.74. A value of φ < 1 might reflect a situation where the host country

government enacts policies to attract FDI (e.g., fiscal or financial incentives, special laws to bypass

cumbersome bureaucratic regulations that domestic firms are ordinarily subjected to), whereas φ > 1

could reflect the usual additional costs of business discussed earlier. Thus, φ = 1.74 would then reflect

costs that are high enough such that the overall efficiency of foreign firms is only 15% greater than that

of domestic firms despite the former having a technological advantage that is “twice” that of the latter.

Under both of these scenarios, H/L ratios are held fixed for each country in the benchmark analysis.

Hence, the resulting differences in the growth rates do not reflect human capital differences, rather

they reflect variations in FDI. Both scenarios are studied separately for the CES and the Cobb Douglas

production function cases, in sections 3.1 and 3.2, respectively.

3.1 CES Production Function

The benchmark results for the CES production function with ρ = 0.2 are reported in table 2. As table

2 indicates, when µ = 0.1, foreign production equals around 6.5% of domestic production, while, when

µ = 0.2, the same ratio increases to around 15.5%. These two values correspond to the share of foreign

production in total production to be 6.1% and 13.4%, respectively.30 Hence, as alluded to earlier, we

use µ = 0.1 in most of the analysis. However, for the sake of completeness, the tables also list results

for increments of 0.1 for µ until µ = 0.6.31

3.1.1 Changes in Relative Productivities and Shares of MNE

The first scenario capturing an increase in the foreign presence is an exogenous increase in the FDI share

(higher µ). Table 2 lists the growth rates for the three different levels of financial development in addition

30The tables report the ratio of foreign production to domestic production, i.e.,Pf Yf

PdYd. Using these share values it is

possible to impute the share of foreign production in total production, i.e.,Pf Yf

PdYd+Pf YF. For example, if foreign production

is 6.5% of domestic production, this corresponds to the share of foreign production in total production being 0.065/(1 +0.065)× 100 = 6.1%

31We restrict our attention to µ ≤ 0.6 since this range covers most realistic values of foreign output shares.

25

to the amount of foreign output relative to domestic output (valued at their respective prices). In order

to ease the discussion, in table 3, we also present the results of table 2 as changes over increments of

0.1 for µ . For example, results in table 3 show that the increase in µ from 0.1 to 0.2 corresponds to a

tripling of the foreign output level. This increase in FDI also creates a 1.25 percentage point increase

in the average growth rate of the financially well-developed countries, a 0.88 percentage point increase

in the average growth rate of the financially medium developed countries, and a 0.61 percentage point

increase in the average growth rate of the financially poorly developed countries. That is, for the same

amount of increase in the share of FDI, the additional growth rates made possible in financially well

developed countries are almost double those made possible in financially poorly developed countries.

These numbers may appear to be quite high and one might wonder if the 1.25 percentage point

increase for developed economies is an overestimate. There are a couple of things to keep in mind.

First, note that we have assumed Af/φAd = 2 in these exercises. For financially developed economies,

the actual gap between domestic and foreign producers is likely to be much lower and thus the estimate

might be too high. Secondly, as µ increases, it is possible that new MNEs entering a domestic market

might be of lower productivity than the first entrants. This could also potentially further reduce the

productivity gap between domestic and foreign firms. Nevertheless, the differences in growth rates

particularly between the medium level and the low level groups is still significantly different.

Another interesting result that emerges from table 3 is that the change in the growth rates are higher

when initial FDI participation is greater. For example, as µ increases from 0.1 to 0.2, the additional

growth for a country with poorly developed financial markets is 0.61% while going from 0.3 to 0.4 leads

to an additional growth rate of 1.25%. Of course, one might wonder what actually happens to foreign

output shares (which are a corollary of changes in µ but are easier to interpret). We already have seen

that the movement from 0.1 to 0.2 leads to an output share increase from 6.1% to 13.4% of GDP. From

the fourth column in table 2, it is easy to see that as µ moves from 0.3 to 0.4, foreign output share goes

from 20.4% to 27%.32 Thus, the increase in output share is slightly lower in the second case, while the

32In column 4 of table 2 relative output at µ = 0.3 is 0.257. This means that the foreign output share is 0.257/1.257 =0.204. Similarly for µ = 0.4, we have 0.369/1.369 = 0.269 .

26

increase in growth rate is higher.

The fact that growth rates go up when the share of foreign output increases is easy to understand in a

scale-effect driven model such as ours. More productive foreign firms raise the scale of the economy and

thus increase the number of varieties which raises the growth rate. The non-linearity is less transparent.

One possible explanation is given by the indirect horizontal spillover this increased share of foreign firms

creates. As the share of foreign firms increase, while the share of domestic firms decreases, the ones that

remain in the latter group benefit from the increased number of varieties and thus also have a higher

level of technology, and this further adds to the growth rate.

Moving on to table 4, we obtain similar qualitative results when the increase in the extent of foreign

presence is captured through an increase in the relative productivity, Af/(φAd). These results, combined

with the ones from the earlier table, suggest that regardless of the source of the increase in the extent

of foreign presence in the local economy, for the same magnitude of increase in foreign presence, the

additional growth effects generated in the local economy are higher for the financially well developed

countries than those generated in the financially medium developed countries, and these are higher than

those generated in the financially poorly developed countries. However, an important difference is that

the additional growth rates generated by improvements in the relative productivity of the foreign firm

are quantitatively much lower than those discussed previously for the case of an increase in the share

of FDI (higher µ). For example, 15% increase in the relative productivity of the foreign firms increases

the growth rate of the financially well developed countries by 0.03 percentage points, the growth rate

of the financially medium developed countries by 0.02 percentage points, and the growth rate of the

financially poorly developed countries by 0.01 percentage points. The higher relative productivity of

the foreign firm corresponds to a only 4.2% increase in the total value of foreign production, pfYf , and

thus only a marginal increase in the share of foreign production in total production. These results hold

qualitatively across alternative µ assumptions.

Obviously, one would be led to wonder why the effects are so dissimilar. An obvious resolution lies in

the way that the two alternative scenarios work. Irrespective of the productivity advantage that foreign

firms enjoy, an increase in µ ensures a higher share of total expenditures will be devoted to the output

27

produced by foreign firms. The fact that Af > Ad ensures that this shift translates into a scale effect.

Thus, the exercise in altering µ, simply answers the question—given realistic productivity differences

between domestic firms and foreign firms—what would a higher share of multinational production mean

for the economy at various levels of financial development? On the other hand, for any given µ, changes

in Af relative to Ad have effects that are slightly more “indirect” in the following sense. An increase

in Af/Ad will reduce the relative price of the foreign good and thus will create a substitution away

from the domestic good towards the foreign good. Thus, while the relative price goes down, the relative

quantity goes up. With the elasticity of substitution being more than 1, we know that the overall

effect is to increase pfYf relative to pdYd. However, as the numbers in table 4 indicate, the changes

are small, and thus the overall growth effect, not surprisingly, will be small. One possibility is that

the choice of ρ = 0.2, which implies an elasticity of substitution of 1. 25, has an important bearing on

these magnitudes. In the next subsection, which deals with the sensitivity of our results, we explore the

implications of varying this parameter.

An alternative way is to compare the elasticities of changes in growth due to changes in the param-

eters of the foreign production firm. For example, instead of restricting ourselves to specific increases

in µ or Af/Ad (which may not be strictly comparable), we could compare the subsequent simultaneous

increases in the share of MNE output in total GDP and the associated increase in the growth rate. To

fix ideas, consider row 1 of both tables 3 and 4. In the case of countries with poorly developed financial

markets, following an increase in µ from 0.1 to 0.2, we see that the rate of growth of GDP increases by

0.61% while the share of MNE output in total GDP increases by 7.3% (from 6.1% to 13.4%). Dividing

the former by the latter produces a value of 0.08. This is equivalent to saying that for every 1% increase

in the share of MNE output in GDP, the growth rate of the economy rises by 0.08%. Now consider

instead an increase in Af/Ad. Begining from the benchmark (row 1 of 2), a 15% increase in Af/Ad,

as we have already seen, raises the growth rate for poorly financially developed countries by 0.01%.

At the same time the share of output in GDP increases from 6.1% to 6.28%—a 0.18% increase in its

share. Here the elasticity is 0.05. This suggests that following an increase in Af/Ad, every 1% increase

in the MNE share of output is associated with a 0.05% increase in the growth rate Thus, the elasticity

28

measures of the effects of changes in µ and Af/Ad are much less disparate. We can also revisit the

comparison between countries with well developed financial markets and countries with poorly devel-

oped financial markets. In the case of the former, the elasticity measures yield values of 0.17 and 0.16

following increases in µ and Af/Ad respectively. Like our earlier findings, we still see that an increase in

MNE share of output is associated with higher rates of economic growth for financially well developed

economies.

Thus far, we have considered two alternative scenarios with qualitatively similar but quantitatively

distinct outcomes. This leads to the next question—which one is more likely to hold in practice?

The first scenario where µ increases seems to be more applicable to a “cross-section” analysis. With

two countries beginning at the same MNE share (of GDP) but different levels of financial market

development, it tells us what happens to the growth rate if the MNE share of GDP increases further.

Alternatively, going down column (3) of table 2—we can ask what happens to the growth rate for

different MNE shares for the same level of financial development. These are also the kind of questions

that growth regressions often seek to answer. The second scenario, where Af increases relative to Ad,

addresses a slightly different issue. It provides a framework to understand what happens as some firms

shift to using a more productive technology. This for instance, would be applicable when domestic

firms are acquired by multinational enterprises, which then bring their superior technology to these

firms. Obviously, this also reflects greater MNE “participation,” however, it does not take an increase

in output share as a given but as an endogenous outcome of this change. Thus, both scenarios have

their respective contributions.

3.1.2 Sensitivity Analysis

Next we examine how the results change with the other parameters or “local conditions.” We focus on

changes in relative skill endowments across countries (varying H/L), the effects of alternative produc-

tivity gaps, changes in the cost of doing business (φ), and finally, varying the elasticity of substitution

(by varying ρ) .

Changes in Labor Endowments: The above exercise kept the relative labor endowments constant

29

across the three groups of countries in order to observe the differences solely on account of financial

market development differences and changes in the share and/or productivity of foreign firms. The

three groups however also differ in their relative labor endowments, as shown in the lower panel of

table 1. When allowing for different labor endowments, table 5 shows that the growth effects of higher

FDI in the countries with well developed financial markets are three times more than the ones with the

poor developed financial markets. Tables 6 and 7 present, respectively, the results when we allow for

the relative labor endowments to differ among the three groups together with changes in the share of

foreign firms, and with changes in the productivity of foreign firms.

When we compare table 5 to table 2, we see that the actual growth rates for countries with medium

and poorly developed financial markets are now even lower. Indeed, the growth rate of the countries

with low levels of financial market development is now only 0.91% compared to 1.42% earlier. Thus,

the introduction of human capital variations across groups exacerbates differences in growth rates.

The incremental effects of changes in µ can be inferred by comparing tables 3 and 6. The differences

in the additional growth rates are also much higher once one allows for human capital differences. When

µ increases from 0.1 to 0.2, the countries with medium level of financial development see their growth

increase by 0.70%, while countries with low level of financial development see their growth increase by

0.43%.

When focusing on productivity gaps, we also find that growth rate differences become larger. Com-

paring the results in tables 4 and 7, one observes that when µ = 0.2, for example, the additional growth

rate generated is 0.06 (table 4) for the financially medium developed countries. When the labor endow-

ments of these countries decrease to the actual level, the additional growth rate decreases to 0.05 (table

7). These results imply that the 0.01 percentage points additional growth was due to the higher quality

labor endowments. However, like before, productivity gaps themselves have growth effects that are of

much smaller magnitude than changes in µ.

Overall, these results suggest that countries with more skill-intensive labor endowments benefit more

in terms of growth effects from FDI, which is consistent with the empirical studies such as Borensztein

et al. (1998).

30

Alternative Measures of Relative Productivity: So far in the analysis, in setting the parameters

regarding the relative productivity of the foreign and domestic firms, Af/φAd, we made use of the

information from macro level studies showing that the relative productivity difference between the in-

dustrialized and developing countries is approximately 2. Micro level studies provide further information

regarding the productivity differences between foreign owned and domestic owned firms and, as we dis-

cussed above, a wide range of micro estimates are available. We report results in table 8 panels A and

B starting with the lowest value from the micro evidence, namely 1.15 (taken from Aitken and Harrison

(1999)) and allow for increments of approximately 15% in this value.

Panel A shows the additional growth rates observed in the three groups of countries when the

technology gap among the foreign and domestic firms change for different values of µ. The results

for the benchmark case (µ = 0.1) suggest that increments of 15% increases in the technology gap

between the foreign and domestic firms creates additional growth rates of 0.020 percentage points in

the financially well developed countries, 0.010 percentage points in the financially medium developed

countries, and 0.006 percentage points in the financially poorly developed countries. If the technology

gap measure increases by 100%, to 2.3, one has to look at the cumulative of the additional growth

values reported in Panel A. For the financially well developed countries, this doubling of the relative

productivity measure creates an additional 0.1 percentage point growth, while creating around 0.05

percentage point growth in the financially medium developed, and around 0.03 percentage point growth

in the financially poorly developed countries. Panel B alternatively looks into the additional growth

rates due to increased foreign presence measured through changes in µ, rather than through changes in

the technology gap. The same results prevail, where the additional growth rates are almost triple for

the financially well developed countries than for the financially poorly developed countries.

Changes in the Cost of Doing Business: While the relative productivity between foreign and do-

mestic firms can change due to the changes in the foreign and the domestic firms’ gross productivity,

an alternative source of change could be alterations in the cost of doing business, φ. The effects of

a reduction of the cost of doing business in our model are similar to those of a relative increase in

the productivity to foreign firms just described. Note that although the interpretation is symmetric,

31

the policy implications are different. One suggests that the authorities should improve the business

environment to benefit more from FDI; the other that attracting more productive foreign firms relative

to domestic firms, everything else being equal, implies higher growth rates.

Changes in the Elasticity of Substitution: Table 9 compares the growth rates in the high, medium

and low financially developed economies for ρ = −0.2 in the upper panel and ρ = 0.2 in the lower

panel. In particular, when the elasticity of substitution is greater than 1 (ρ > 0), for the same value

of µ, we observe much higher growth rates. Further increases in µ (i.e., a greater MNE presence),

while leading to increases in the growth rate for ρ = 0.2, actually reduces growth rates when ρ = −0.2,

despite increasing MNE share in output. Thus, clearly, the elasticity of substitution in the aggregator

plays a key role in our numerical exercises. These results suggest one must be cautious when talking of

attracting “FDI that is complementary to local production.” Such complementarity is useful when one

talks of final and intermediate industry relationships. However, it does not necessarily raise the growth

rates when domestic and foreign producers supply complementary final goods.

Finally, we consider the extent to which a change in the growth rate following an increase in the

overall productivity gap (Af/φAd) is affected by the choice of the elasticity of substitution parameter.

As earlier, we consider the implications of a 15% increase in the overall productivity gap. Figures

1 and 2 show the non-monotonic relationship between ρ and the additional growth rates created by

increased FDI for the financially developed economies. Figure 1 depicts the relationship when ρ > 0,

i.e., the elasticity of substitution is greater than 1. Figure 2 depicts the relationship when the elasticity

of substitution is less than 1 (ρ < 0).33 We also consider the effects for various values of µ. Beginning

with figure 1, we see that as Yf and Yd become more substitutable (ρ increases), the additional growth

generated actually declines. This is true for all values of µ considered here. This might initially seem

counter-intuitive. However, note that when the elasticity of substitution is very high, combined with the

fact that Yf already uses technology that is twice as productive and µ is fixed, optimal allocation would

cause domestic output to have already been substituted by foreign output as much as possible. Thus,

33The case where ρ = 1 is the Cobb Douglas case discussed in the next section. For values of ρ in the neighborhood of1, there is a sharp discontinuity, and hence we do not include ρ = 1 in these diagrams.

32

further increases in Af create only limited additional substitution possibilities and hence the additional

growth effects are small.

In the situation where the two products are more complementary (i.e., ρ < 0), we see again that

increases in the elasticity of substitution (i.e., as the absolute value of ρ falls) leads to lower additional

growth rates. However, note that the effects are actually even smaller here. The overall impression one

can draw is that the introduction of a more advanced technology by the MNE while raising the growth

rate, seems to be less effective when a) ρ < 0 and b) the products become more substitutable. The

figures furthermore show that these results remain unchanged across alternative initial values of foreign

presence in the market, i.e., for alternative values of µ.34

3.2 Cobb Douglas production function

This section discusses the results for the Cobb Douglas production function. As seen in table 10,

column (1), there are large differences in growth rates across groups only due to differences in the level

of development of the financial markets. Next, we consider the effect of increasing Af by 25% as shown

in table 10, column (2). Comparing columns (1) and (2) shows very little change in growth rates. The

growth rates are slightly higher for the high and medium financially developed countries (i.e., 2.49 vs.

2.52; 3.61 vs. 3.65). Thus, it seems that the marginal effect of raising Af is very small. This was true for

the CES case and carries over here as well. With foreign firms producing only 5% of the total output,

it would be unreasonable to expect improvements in their productivity to have large measurable effects

on the growth rate of the economy.

In column (3), we allow human capital ratios to vary across country groups. Everything else is as

in column (1). Comparing to column (1), we see that the growth rates for medium and low financially

developed countries decrease, whereas the rates for the high financially developed country are the same

since the ratio for the latter is unchanged. In column (3), the countries with high levels of financial

development grow twice as fast as the countries with low levels of financial development. In Column

34The figures and table 9 use benchmark parameters. In particular, the ratio of skilled to unskilled human capital isconstant across groups.

33

(4) we again increase Af by 25%, except that now human capital ratios vary across the three groups.

Again as in the case of comparing column (1) and column (2), the differences between (3) and (4) are

negligible.

In column (5), we decrease the share of the domestic firm in total output to 75% (i.e. η = 0.75)

thus taking the share of the foreign firms up to 25%. This probably represents an upper bound in terms

of foreign ownership. Everything else is as in column (3). Growth rates are lower for all the levels

of financial development but the qualitative results are the same. Low financially developed countries

grow at less than one third of the speed of high financially developed countries. Column (6) repeats the

same exercise with the exogenous increase in productivity of MNE. The growth rates are higher again,

though more so than the previous cases, where the increase was negligible. These findings are parallel

to the findings reported for the CES case, where the magnitude of effects are much larger when the

implicit share of foreign production is higher in the domestic economy. The bottom line is that the role

of financial markets is extremely important in realizing the growth effects of higher FDI.

4 Conclusions

Although there is a widespread belief among policymakers that FDI generates positive productivity

externalities for host countries, the empirical evidence fails to confirm this belief. In the particular case

of developing countries, both the micro and macro empirical literatures consistently finds either no effect

of FDI on host countries firms productivity and/or aggregate growth or negative effects. The theoretical

models of FDI, on the other hand, imply that FDI is beneficial for the host country’s development.

In this paper, we try to bridge this gap between the theoretical and the empirical literatures. The

model rests on a mechanism that emphasizes the role of local financial markets in enabling FDI to

promote growth through the creation of backward linkages. When financial markets are developed

enough, the host country benefits from the backward linkages between the foreign and domestic firms

with positive spillovers to the rest of the economy.

Our calibration exercises show that an increase in FDI leads to higher growth rates in financially

34

developed countries compared to those observed in financially poorly-developed ones. Moreover, the

calibration section highlights the importance of local conditions (absorptive capacities) for the effect of

FDI on economic growth. We find larger growth effects when goods produced by domestic firms and

MNEs are substitutes rather than complements. Policymakers should be cautious when implementing

policies aimed at attracting FDI that is complementary to local production. Desired complementarities

are those between final and intermediate industry sectors; not necessarily between domestic and foreign

final good produces. Finally, by varying the relative skill ratios—while assuming that MNEs use skilled

labor more intensively—our results highlight the critical role of human capital in allowing growth benefits

from FDI to materialize.

Some caveats are in order. We have focussed on only one kind of spillover. There are likely to be

additional spillovers and technology transfers. Besides, our results are based on a model that takes FDI

as given. The decision of a firm to outsource or invest abroad (and the potential to generate linkages)

may depend on the conditions of the country and on the characteristics of the firm.35

35See Antras and Helpman (2004).

35

5 References

Aghion, P., D. Comin, and P. Howitt, 2006. “When Does Domestic Saving Matter for EconomicGrowth?” NBER Working Paper 12275.

Aghion, P., P. Howitt, and D. Mayer-Foulkes, 2005. “The Effect of Financial Development onConvergence: Theory and Evidence,” Quarterly Journal of Economics 120, 173-222.

Aitken, B. J. and A. Harrison, 1999. “Do Domestic Firms Benefit from Direct Foreign Invest-ment? Evidence from Venezuela,” American Economic Review 89, 605-618.

Alfaro, L., A. Chanda, S. Kalemli-Ozcan, and S. Sayek, 2004. “FDI and Economic Growth, TheRole of Local Financial Markets,” Journal of International Economics 64, 113-134.

Alfaro, L. and A. Rodriguez-Clare, 2004. “Multinationals and Linkages: Evidence from LatinAmerica,” Economia 4, 113-170.

Antras, P., 2003. “Firms, Contracts and Trade Structure,” Quarterly Journal of Economics 118,1374-1418.

Antras, P. and E. Helpman, 2004. “Global Sourcing,” Journal of Political Economy 112, 552-580.

Barba Navaretti, G. and A. Venables, 2004. Multinational Firms in the World Economy. Prince-ton: Princeton University Press.

Basu, S., 1996. “Procyclical Productivity: Increasing Returns or Cyclical Utilization,” QuarterlyJournal of Economics 111, 719-751.

Blomstrom, M., 1986. “Foreign Investment and Productivity Efficiency: The Case of Mexico,” Jour-nal of International Economics 15, 97-110.

Blomstrom, M. and A. Kokko, 1998. “Multinational Corporations and Spillovers,” Journal of Eco-nomic Surveys 12, 247-277.

Borensztein, E., J. De Gregorio, and J-W. Lee, 1998. “How Does Foreign Direct Investment Af-fect Economic Growth?” Journal of International Economics 45, 115-135.

Burnstein, A. and A. Monge-Naranjo, 2005. “Aggregate Consequences of International Firms inDeveloping Countries,” mimeo.

Carkovic, M. and R. Levine, 2002. “Does Foreign Direct Investment Accelerate Economic Growth?”University of Minnesota, Working Paper.

Caves, R., 1974. “Multinational Firms, Competition and Productivity in the Host Country,” Eco-nomica 41, 176-193.

Duffy, J., C. Papageorgiou, and F. Perez-Sebastian, 2004. “Capital-Skill Complementarity? Ev-idence From a Panel of Countries,” The Review of Economics and Statistics 86, 327-344.

Dunning, J. H., 1981. International Production and the Multinational Enterprise. London: GeorgeAllen and Unwin.

Durham, K. B., 2004. “Absorptive Capacity and the Effects of Foreign Direct Investment and EquityForeign Portfolio Investment on Economic Growth,” European Economic Review 48, 285-306.

36

Edwards, S. and C. Vegh, 1997. “Banks and Macroeconomic Disturbances under PredeterminedExchange Rates,” Journal of Monetary Economics 40, 239-278.

Erosa, A., 2001. “Financial Intermediation and Occupational Choice in Development,” Review ofEconomic Dynamics 4, 303-334.

Ethier, W., 1982. “National and International Returns to Scale in the Modern Theory of Interna-tional Trade,” American Economic Review 72, 389-405.

Gao, T., 2005. “Foreign Direct Investment and Growth Under Economic Integration,” Journal ofInternational Economics 67, 157-174.

Gollin, D., 2002. “Getting Income Shares Right,” Journal of Political Economy 110, 458-474.

Goldsmith, R. W., 1969. Financial Structure and Development. New Haven: Yale University Press.

Gorg, H. and D. Greenway, 2004. “Much ado about nothing? Do domestic firms really benefitfrom foreign direct investment?” World Bank Research Observer 19, 171-197.

Gorg, H. and E. Strobl, 2002. “Multinational Companies and Indigenous Development: An Em-pirical Analysis,” European Economic Review 46, 1305-1322

Graham, E. and P. Krugman, 1991. Foreign Direct Investment in the United States. WashingtonD.C.: Institute for International Economics.

Greenwood, J. and B. Jovanovic, 1990. “Financial Development, Growth and the Distribution ofIncome,” Journal of Political Economy 98, 1076-1107.

Grossman, G. and E. Helpman, 1990. “Comparative Advantage and Long-Run Growth,” Ameri-can Economic Review 80, 796-815.

Grossman, G. and E. Helpman, 1991. Innovation and Growth in the Global Economy . Cambridge,MA: MIT Press.

Hall, R. E. and C. Jones, 1999. “Why Do Some Countries Produce So Much More Output perWorker than Others?” The Quarterly Journal of Economics 114, 83-116.

Haskel, J. E., S. C. Pereira, and M. J. Slaughter, 2002. “Does Inward Foreign Direct Invest-ment Boost the Productivity of Local Firms?” NBER Working Paper 8433.

Helpman, E., 1984. “A Simple Theory of International Trade and Multinational Corporations,”Journal of Political Economy 92, 451-471.

Helpman, E. and P. Krugman, 1985. Market Structure and Foreign Trade: Increasing Returns,Imperfect Competition, and the International Economy. Cambridge, MA: MIT Press.

Helpman, E., M. Melitz, and S. R. Yeaple, 2004. “Export versus FDI with Heterogeneous Firms,”American Economic Review 94, 300-316.

Hermes, N. and R. Lensink, 2003. “Foreign Direct Investment, Financial Development and Eco-nomic Growth,” Journal of Development Studies 40, 142-163.

Hirschman, A., 1958. The Strategy of Economic Development. New Haven: Yale University Press.

Hobday, M. 1995. Innovation in East Asia: The Challenge to Japan. Cheltenham: Edward Elgar.

37

Javorcik, B. S., 2004. “Does Foreign Direct Investment Increase the Productivity of Domestic Firms?In Search of Spillovers Through Backward Linkages,” American Economic Review 94, 605-627.

Jones, Charles I., 1995. “R & D-Based Models of Economic Growth,” The Journal of Political Econ-omy 103, 759-784.

Kindleberger, C. P., 1969. American Business Abroad. New Haven: Yale University Press.

King, R. and R. Levine, 1993a. “Finance and Growth: Schumpeter Might be Right,” QuarterlyJournal of Economics 108, 717-738.

King, R. and R. Levine, 1993b. “Finance, Entrepreneurship and Growth: Theory and Evidence,”Journal of Monetary Economics 32, 513-542.

Kugler, M., 2006. “Spillovers from Foreign Direct Investment: Within or Between Industries,” Jour-nal of Development Economics, forthcoming.

Levine, R., N. Loayza, and T. Beck, 2000. “Financial Intermediation and Growth: Causality andCauses,” Journal of Monetary Economics 46, 31-77.

Lin P. and K. Saggi, 2006. “Multinational Firms, Exclusivity, and Backward Linkages,” Journal ofInternational Economics, forthcoming.

Lipsey, R. E., 2002. “Home and Host Country Effects of FDI,” NBER Working Paper 9293.

Lipsey, R. E., 2003. “Foreign Direct Investment and the Operations of Multinational Firms: Con-cepts, History, and Data,” in E. Kwan Choi and James Harrigan (eds.) Handbook of InternationalTrade. Oxford: Blackwell.

Markusen, J. and A. J. Venables, 1999. “Foreign Direct Investment as a Catalyst for IndustrialDevelopment,” European Economic Review 43, 335-338.

Markusen, J., 1984. “Multinationals, Multi-Plant Economies and the Gains from Trade,” Journal ofInternational Economics 16, 205-226.

Mataloni, R. J., 2005. “U.S. multinational companies: operations in 2003,”Survey of Current Busi-ness July, 9-29.

McKinnon, R., 1973. Money and Capital in Economic Development. Washington: Brookings Insti-tute.

Rodrıguez-Clare, A., 1996. “Multinationals, Linkages and Economic Development,” American Eco-nomic Review 86, 852-873.

Ruhl, K., 2005. The Elasticity Puzzle in International Economics, mimeo.

Weil, D., 2004. Economic Growth. Boston: Addison-Wesley.

Xu, B., 2000. “Multinational Enterprises, Technology Diffusion, and Host Country Productivity Growth,”Journal of Development Economics 62, 477-493.

38

A Modelling Financial Markets

We present a bare-bones model of imperfect financial markets using the costly state verification ap-proach. The model is adapted from King and Levine (1993b).36 As in their model, we assume individ-uals have equal financial wealth which is a claim on profits of a diversified portfolio of firms engagedin innovative activity. Some individuals do have the ability to manage innovation, but this does notlead them to accumulate different levels of wealth from the rest of individuals in the economy. Thesepotential entrepreneurs have the ability to successfully manage a project with probability ψ. Theseabilities are unobservable to both the entrepreneur and the financial intermediary. However, the actualcapability of such an individual to manage a project can be ascertained at a cost, F .37 The main twodifferences are the following. First, consistent with our model, upfront investments require capital in-stead of labor. The second difference is related to our assumption regarding the structure of verificationcosts. We assume these costs to be proportional to the set up costs for any project. The main advantageof this approach is that it allows us to retain the balanced growth properties of the model while allowingin principle to make total verification costs decreasing in the level of overall technology and hence in thelevel of development.38 Therefore, one could argue that our setup automatically relates more efficientfinancial markets to higher levels of development.39

We depart from the main text in that now innovation and imitation projects are potentially risky andthere is a probability ψ of the project being run successfully. Which potential entrepreneur will managea project successfully is unknown both to the entrepreneur and the intermediary. The intermediary canspend an amount F to reduce the uncertainty regarding the project’s outcome. Further, we postulatethe following structure on F,

F = fra

nθ. (40)

Therefore, since setup costs require anθ units of K, then the cost of verification simply is proportional

to total setup costs and f represents that factor of proportionality. While not necessary for our model,it seems intuitive that f should be less than 1—verification costs are likely to be lower than setup costs.

If the value of a successful project is q, with a competitive intermediation sector, in equilibrium wemust have ψq = F.40 The value of a successful project is simply the present discounted value of profits(v) minus the set up costs (η). Therefore, the above condition can be rewritten as, ψ(v − η) = F .Further, from our model we had setup costs for each blueprint to be ra

nθ ,

⇒ ψ(v − ra

nθ

)= F. (41)

The standard arbitrage condition from equation (20) continues to hold,

36See Aghion, Howitt and Mayer-Foulkes (2005) for an alternative modelling strategy for imperfect financial markets inan endogenous growth model. In their model, entrepreneurs can pay an upfront cost and defraud the lenders. Incentivecompatibility constraints and arbitrage conditions then leads to an upper bound on how much is actually invested in newinnovations which is lower than the optimal investment amount. The gap between the two is an inverse function of thedegree of creditor protection. Their model seems more suited to “quality ladder” Schumpeterian models, whereas the Kingand Levine structure is more easily incorporated into a product variety setup such as ours.

37Obviously entrepreneurs cannot evaluate themselves and credibly communicate the results to others.38Indeed King and Levine (1993b, 518) suggest this modification as a potentially useful extension of their model.39The objective of our paper, as mentioned, is not to model the relation between financial markets and development but

instead the role of financial markets in allowing an economy to reap the benefits of potential FDI spillovers.40Of course in the background we assume that for an intermediary it is better to do this evaluation rather than simply

lending the money and not incurring the verification cost.

39

π

v+v

v= r. (42)

Substituting equation (40) in equation (41), we continue to get as in the main model vv = −θ nn . Wecan combine this with the previous equation to get, πv − θ

nn = r. In our model the per intermediate firm

operating profit was (see equation (19)), πi = (1−α)n [λpdYd + λpfYf ] . Noting that v = f

ξ ranθ + ra

nθ , andusing Yi = Yi/n

1−θ, we can obtain an expression for the growth rate similar to the one derived in themain text,

g =n

n=λ

θ

(1− τ) (1− α)

ra(fψ + 1

) [λpdYd + λpdYf

]− r

θ.

Of course, what was earlier represented in the main text by i can now be substituted by r(fψ + 1

).

In practice, fψ is likely to be unobservable across countries. Therefore, in the numerical exercises, when

we use the spread, theoretically, we are measuring r fψ . Thus, a higher spread has the same effect asa higher verification cost. Further, for every unique f, given r and ψ, there is a unique value of thespread. Therefore, using the spread between lending and borrowing rates serves as a convenient proxyfor verification costs.

B Solving the Model with a CES Aggregator

As mentioned in the text, we begin with six equations,

pd =A−1d β−βd

d γ−γdd

λλ

(δ−δ(1− δ)−(1−δ)

α

)λwβd+δλu wγd+(1−δ)λ

s , (43)

pf =φA−1

f β−βf

f γ−γf

f

λλ

(δ−δ(1− δ)−(1−δ)

α

)λ

wβf+δλu w

γf+(1−δ)λs , (44)

pfpd

= µ

[Yd

Yf

]1−ρ

, (45)

pd =(1− µεp1−ε

f

)⇒ pd = pd (pf ) , (46)

(βd + δαλ) pdYdwu

+(βf + δαλ) pf Yf

wu= L, (47)

(γd + (1− δ)αλ) pdYdws

+(γf + (1− δ)αλ) pf Yf

ws= H. (48)

Also, recall that the growth rate of varieties is pinned down by equation (26),

n

n=

(1− α)λθia

[pdYd + pf Yf

]− r

θ. (49)

We next list the steps involved in arriving at a solution for this setup:

40

1) First of all note that we can use equations (43) and (44) to express efficiency wages as a functionof the prices of foreign and domestic goods,

wu = Adu∆Afup

(γf +(1−δ)λ)(γf−γd)

d p

−(γd+(1−δ)λ)

(γf−γd)f , (50)

ws = Ads∆Afsp

−(βf +δλ)(γf−γd)d p

βd+δλ

(γf−γd)f . (51)

From these two equations we get,

pdpf

=A−1d β−βd

d γ−γdd

φA−1f β

−βf

f γ−γf

fs

(wuws

)γf−γd

,

pdpf

=ΛfΛd

(wuws

)γf−γd

. (52)

where Λd =(Adβ

βdd γ

γdd

),Λf =

(Afβ

βf

f γγf

f

)/φ. Dividing equation (47) by (48),

⇒(βd + δαλ) + (βf + δαλ) pf Yf

pdYd

(γd + (1− δ)αλ) + (γf + (1− δ)αλ) pf Yf

pdYd

=wuL

wsH. (53)

Note that the cost minimization equation (45) can be rewritten as,[1µ

pfpd

] −11−ρ

=Yf

Yd. (54)

Substituting this into equation (53),

(βd + δαλ) + (βf + δαλ)(pf

pd

)1− 11−ρ[

1µ

] −11−ρ

(γd + (1− δ)αλ) + (γf + (1− δ)αλ)(pf

pd

)1− 11−ρ[

1µ

] −11−ρ

=wuL

wsH.

Further using equation (52),

(βd + δαλ) + (βf + δαλ)(pf

pd

)1− 11−ρ[

1µ

] −11−ρ

(γd + (1− δ)αλ) + (γf + (1− δ)αλ)(pf

pd

)1− 11−ρ[

1µ

] −11−ρ

=(

ΛdΛf

pdpf

) 1γf−γd L

H.

Finally using equation (5),(1− µεp1−ε

f

) 11−ε = pd, we can rewrite the expression above to obtain,

41

⇒(βd + δαλ) + (βf + δαλ)

(pf

(1−µεp1−εf )

11−ε

)1− 11−ρ [

1µ

] −11−ρ

(γd + (1− δ)αλ) + (γf + (1− δ)αλ)

(pf

(1−µεp1−εf )

11−ε

)1− 11−ρ [

1µ

] −11−ρ

=

1Λ

(1− µεp1−ε

f

) 11−ε

pf

1

γf−γd

L

H

Thus solving for pf = p∗f , where ∗ denotes the solved value.

2) Given pf , we can again use p∗d =(1− µεp1−ε

f

) 11−ε to back out pd.

3) Since we now have both p∗d and p∗f , we can also derive the efficiency wages and the relativeoutputs. To derive the efficiency wages, we can substitute prices into equations (50) and (51), as torewrite them such that we have w∗u = w∗u(pd, pf ) and w∗s = w∗s(pd, pf ). More explicitly, after sometedious rearrangements, we get,

w∗u = Adu∆Afu (p∗d)(γf +(1−δ)λ)

(γf−γd) (p∗f)−(γd+(1−δ)λ)

(γf−γd) , (55)

w∗s = Ads∆Afs (p∗d)

−(βf +δλ)(γf−γd) (p∗f) βd+δλ

(γf−γd) , (56)

where,

Adu =(Adβ

βdd γ

γdd

)(γf +(1−δ)λ)(γf−γd) ; Afu =

(Afβ

βff γ

γff

φ

)−(γd+(1−δ)λ)

(γf−γd);

Ads =(Adβ

βdd γ

γdd

)−(βf +δλ)(γf−γd) ; Afs =

(Afβ

βff γ

γff

φ

) βd+δλ

(γf−γd);

∆ =

(δ−δ(1− δ)−(1−δ)

α

)−λ.

This allows us also to derive the relative wages,

wuws

=

(ΛdΛf

p∗dp∗f

) 1γf−γd

.

From equation (54), we obtain a value for relative outputs,

⇒Yf

Yd=[

1µ

p∗fp∗d

] −11−ρ

. (57)

4) We can write Yf = Y(Yd

).

5) Taking the unskilled labor market equation (47),

(βd + δαλ) p∗dYd + (βf + δαλ) p∗f Yf = w∗uL,

42

6) We can now substitute this into equation (57) and get Y ∗f .7) Thus, we can now derive the growth rate from equation (49):

n

n=

(1− α)λθia

[p∗dY

∗d + p∗f Y

∗f

]− r

θ.

C Solving for the Cobb Douglas Case

Similar to the CES case, we begin with an analogous set of six equations,

pd =A−1d β−βd

d γ−γdd

λλ

(δ−δ(1− δ)−(1−δ)

α

)λwβd+δλu wγd+(1−δ)λ

s , (58)

pf =φ(τ)A−1

f β−βf

f γ−γf

f

λλ

(δ−δ(1− δ)−(1−δ)

α

)λ

wβf+δλu w

γf+(1−δ)λs , (59)

pdpf

=η

1− η

Yf

Yd, (60)

pd = η (1− η)(1−η)

η p−(1−η)

η

f , (61)

(βd + δαλ) pdYdwu

+(βf + δαλ) pf Yf

wu= L, (62)

(γd + (1− δ)αλ) pdYdws

+(γf + (1− δ)αλ) pf Yf

ws= H. (63)

We have six equations and six unknowns. Substituting (60) into equations (62) and (63), we obtain,

(η (βd + δαλ) + (βf + δαλ)

(1− η)

)pf Yf = wuL, (64)(

η (γd + (1− δ)αλ) + (1− η) (γf + (1− δ)αλ)(1− η)

)pf Yf = wsH. (65)

Therefore, we can derive the wage premium in this setup,

⇒ wswu

=η (γd + (1− δ)αλ) + (1− η) (γf + (1− δ)αλ)

η (βd + δαλ) + (1− η) (βf + δαλ)L

H. (66)

At the same time note that the growth equation,

(1− α)λia

[pdYd + pf Yf

]− θ

n

n= r ⇒

((1− α)λia (1− η)

)pf Yf − θ

n

n= r. (67)

To solve for the endogenous rate of growth of varieties we simply need to figure out pf Yf . This in

43

turn requires us to figure out either wsH, (equation (65)), or wuL, (equation (64))41. If we proceedwith the former, we have,(

(1− α)λia

)(wsH

η (γd + (1− δ)αλ) + (1− η) (γf + (1− δ)αλ)

)− θ

n

n= r. (68)

Similarly, to solve for ws and wu, equations (58) and (59) can be rewritten such that we have,wu = wu(pd, pf ) and ws = ws(pd, pf ). More explicitly, after some tedious rearrangements, we get,

wu = Adu∆Afup

(γf +(1−δ)λ)(γf−γd)

d p

−(γd+(1−δ)λ)

(γf−γd)f , (69)

ws = Ads∆Afsp

−(βf +δλ)(γf−γd)d p

βd+δλ

(γf−γd)f . (70)

Note that these are exactly the same as the corresponding CES equations (55) and (56). We canuse these expressions for efficiency wages and substitute them into equation (66) and write prices as afunction of L/H. Again this involves some tedious algebra but ultimately gives us

pfpd

=φAdβ

βdd γ

γdd

Afββf

f γγf

f

(η (γd + (1− δ)αλ) + (1− η) (γf + (1− δ)αλ)

η (βd + δαλ) + (1− η) (βf + δαλ)L

H

)γf−γd

. (71)

Therefore, we have the relative prices completely in terms of exogenous variables. As expected, theprices are inversely related to the relative TFP’s of the sectors. Further as long as γf > γd (share ofH is greater in the foreign sector), a decrease in the L/H ratio leads to a decrease in relative prices.As human capital becomes relatively more abundant, the sector that uses this factor more intensivelybenefits more from the lower cost and therefore charges a lower price. Finally we can use equation (61)in conjunction with (71) to solve explicitly for pf and pd. Once we have these two solutions, we cansubstitute them back into equations (69) and (70) and derive the explicit values for wu and ws. All ofthis involves another round of tedious algebra, and we get the following solutions,

pf = Υ(

ΛdΛf

)η (ΦL

H

)η(γf−γd), (72)

pd = Υ(

ΛdΛf

)−(1−η)(ΦL

H

)−(γf−γd)(1−η), (73)

wu = ∆ΥΛηdΛ1−ηf

(Φ−1H

L

)Ψγ

, (74)

ws = ∆ΥΛηdΛ1−ηf

(ΦL

H

)Ψβ

, (75)

where Υ = ηη (1− η)(1−η) ; Λd =(Adβ

βdd γ

γdd

); Λf =

(Afβ

βf

f γγf

f

)/φ; Ψβ = η (βd + δλ)+(1− η) (βf + δλ) ;

Φ =η(γd+(1−δ)αλ)+(1−η)(γf+(1−δ)αλ)

η(βd+δαλ)+(1−η)(βf+δαλ) ; Ψγ = η (γd + (1− δ)λ) + (1− η) (γf + (1− δ)λ) ; (note that Ψβ +

41Given the symmetric nature of Cobb Douglas production functions, ultimately it does not matter which one we proceedwith.

44

Ψγ = 1).To derive the growth rate, we substitute wsH = ∆ΥΛηdΛ

1−ηf (ΦL)Ψβ HΨγ into equation (68) to derive

nn .

D Domestic and Foreign Production as Perfect Substitutes

For this section, note that both firms have to sell the products at the same price and thus we ignorethe aggregator. As mentioned in the main text, we normalize this price to 1. Therefore, Y = Yd + Yf .Working through the model and solving it the same way as in the Cobb Douglas case, we have the totalvalue of output produced by MNEs as,

Yf =(βd + λδα) wsH − (γd + λ (1− δ)α) wuL

(1− λ+ αλ) (βd − βf ), (76)

and the following expression for the domestic production,

Yd =(γf + αλ (1− δ)) wuL− (βf + λδα) wsH

(1− λ+ αλ) (βd − βf ), (77)

where the equilibrium factor prices (which can be derived using cost functions and labor market clearingconditions like before) are given by,

wu =

(δδ(1− δ)(1−δ)

α

)λ 1(Adβ

βdd γ

γdd λ

λ)

γf +(1−δ)λ

γd−γf

Afββf

f γγf

f λλ

φ

(γd+(1−δ)λ)

γd−γf

, (78)

ws =

(δδ(1− δ)(1−δ)

α

)λ (Adβ

βdd γ

γdd λ

λ)(βf +δλ)

γd−γf

φ

Afββf

f γγf

f λλ

βd+δλ

γd−γf

. (79)

While we do not go into the details, observe that for both Yf and Yd to be positive, the parametersmust satisfy certain conditions. Summing the two equations, (76) and (77), and using the fact that(γf − γd) = (βd − βf ) , we can rewrite total final good production in the economy as,

Yd + Yf =(γf − γd) wuL+ (βd − βf ) wsH

(1− λ+ αλ) (βd − βf )=

wuL+ wsH

(1− λ+ αλ). (80)

Substituting equation (80) into the arbitrage condition (26) gives us the equilibrium growth rate forvarieties,

n

n=λ

θ

(1− α)ia

[(wuL+ wsH)(1− λ+ αλ)

]− r

θ. (81)

Improvements in the level of financial market development continue to have clear positive effects onthe growth rate. How does a change in Af (or φ) affect the growth rate of the economy? In order toperform this exercise, we need to solve for the effect of changes in Af on wu and ws. Looking at thereduced form factor price equations, an increase in Af raises the skilled wage per efficiency unit andreduces the unskilled wage per efficiency unit as long as multinationals use skills more intensively, thatis γf > γd. This is because an improvement in the technology of the skill intensive sector raises thedemand for skills more than it raises the demand for raw labor. As a result, this creates an upwardpressure on skilled labor wages. This creates an excess supply of unskilled labor since domestic firms use

45

this kind of labor relatively intensively. As a result, skilled labor wages rise and unskilled labor wagesfall. This suggests that the overall effect on growth can be ambiguous. Moreover, the effect of changesin Af depends also upon the relative stocks of L and H in the economy. If the increase in skilled laborwage bill, more than compensates the reduction in unskilled labor wage bill then the growth rate ofthe economy will go up. Thus, even the skill intensive nature of FDI is not sufficient to ensure thatmore FDI leads to higher growth rates in the economy. If it does raise the growth rate then clearlyboth sectors experience increases in growth rates. This would then be the case of a beneficial spillovereffect. On the other hand, if the increase in skilled wages does not compensate for the reduction inunskilled wages, then the growth rates will diminish. In this case, FDI would have a negative impactin the economy.42 Of course the opposite happens if Ad increases.

42Rodriguez-Clare (1996) and Markusen and Venables (1999) also get ambiguous results in terms of the effects ofmultinationals in the domestic economy stemming from reduction in the demand for local inputs due, for example, to thefact that foreign firms may import intermediate inputs.

46

Table 1: Parameters

Benchmark Parameters

Common parameters for three groupsα = 0.91 r = 0.05 φ = 1

Production Function Parametersβd = 0.34 βf = 0.27 γ = 0.5γd = 0.33 γf = 0.40 Af/Ad = 2µ = 0.1 ρ = 0.2

Group Specific ParametersFinancial Dev. L/H

High (rich) 0.045 5Medium (middle) 0.085 5Low (poor) 0.145 5

Robustness Parameters

Production Function Parametersρ = 0.2 µ = 0.2 Af/Ad = 2

Group Specific ParametersL/H

High (rich) 5Medium (middle) 9Low (poor) 12

Notes: We group countries based on their financial market development levels, using the interest ratespreads. The average spread for the low financially developed (poor) countries, medium financiallydeveloped (middle income) countries and the high financially developed (rich) countries between 2000and 2003 are 14.5%, 8.5%, and 4.5%, respectively. In the benchmark case, all countries have the sameratio of unskilled to skilled labor equal to 5. In the sensitivity analysis, we set the ratio of unskilledlabor to skilled labor equal to 12 for the poor countries, 9 for the middle income countries, and 5 forthe rich countries (taken from Duffy et al. (2004)).

47

Table 2: Benchmark Results

Growth Rate Growth Rate Growth Rate RelativeHigh Financial Medium Financial Low Financial Output

µ Development Development Development (Y f/Y d)

0.1 3.10 2.13 1.42 0.0650.2 4.35 3.01 2.03 0.1550.3 6.17 4.29 2.92 0.2570.4 8.74 6.10 4.17 0.3690.5 12.25 8.57 5.88 0.4870.6 16.97 11.89 8.18 0.612

Notes: See table 1 for the parameter values. Growth rates are in percent. Relative outputs are valuedat their respective price. The relative labor endowments are constant at the level of rich (high financialdevelopment) countries and ρ = 0.2.

Table 3: Increasing Foreign Presence, Changing µ

∆ Growth ∆ Growth ∆ Growth ∆ Relative PercentHigh Financial Medium Financial Low Financial Output Change in

∆µ Development Development Development ∆(Y f/Y d) Y f

0.1 to 0.2 1.25 0.88 0.61 0.09 203.20.2 to 0.3 1.83 1.29 0.89 0.10 114.10.3 to 0.4 2.56 1.80 1.25 0.11 84.80.4 to 0.5 3.51 2.47 1.71 0.12 69.60.5 to 0.6 4.72 3.32 2.30 0.12 59.9

Notes: See table 1 for the parameter values. All changes are in percentage points unless reportedotherwise. Relative outputs are valued at their respective price. The relative labor endowments areconstant at the level of rich (high financial development) countries and ρ = 0.2.

48

Table 4: Increasing Foreign Presence via Increasing MNE Productivity: Af/Ad ↑ by 15%

∆ Growth ∆ Growth ∆ Growth ∆ Relative PercentHigh Financial Medium Financial Low Financial Output Change in

µ Development Development Development ∆(Y f/Y d) Y f

0.1 0.03 0.02 0.01 0.002 4.20.2 0.09 0.06 0.04 0.006 5.00.3 0.19 0.13 0.09 0.009 5.80.4 0.35 0.24 0.17 0.013 6.60.5 0.59 0.41 0.29 0.017 7.20.6 0.94 0.66 0.46 0.022 7.8

Notes: See table 1 for the parameter values. All changes are in percentage points unless reportedotherwise. Relative outputs are valued at their respective price. The relative labor endowments areconstant at the level of rich (high financial development) countries and ρ = 0.2. A 15% increase inAf/Ad implies that this ratio increases in value from 2 to 2.3.

Table 5: L/H Varies by Group

Growth Rate Growth Rate Growth Rate Relative Relative RelativeHigh Financial Medium Financial Low Financial Output Output Output

µ Development Development Development High Medium Low

0.1 3.10 1.68 0.97 0.065 0.065 0.0640.2 4.35 2.38 1.40 0.155 0.153 0.1530.3 6.17 3.40 2.02 0.257 0.255 0.2530.4 8.74 4.84 2.90 0.369 0.365 0.3630.5 12.25 6.79 4.09 0.487 0.482 0.480

Notes: See table 1 for the parameter values. Growth rates are in percent. Relative outputs are valuedat their respective price. The relative labor endowments change together with financial development ashigh, medium and low; ρ = 0.2.

49

Table 6: Increasing Foreign Presence (Changing µ) and L/H varies by Group

∆ ∆ ∆ ∆ ∆ ∆ % ∆ % ∆ % ∆Growth Growth Growth Y f/Y d Y f/Y d Y f/Y d Y f Y f Y f

∆µ High Medium Low High Medium Low High Medium Low

0.1 to 0.2 1.25 0.70 0.43 0.09 0.09 0.09 203.2 202.5 202.20.2 to 0.3 1.83 1.02 0.62 0.10 0.10 0.10 114.1 113.6 113.40.3 to 0.4 2.56 1.43 0.87 0.11 0.19 0.11 84.8 84.4 84.20.4 to 0.5 3.51 1.96 1.19 0.12 0.12 0.12 69.6 69.2 69.1

Notes: See table 1 for the parameter values. All changes are in percentage points unless reportedotherwise. Relative outputs are valued at their respective price. The relative labor endowments changetogether with financial development as high, medium, low; and ρ = 0.2.

Table 7: Increasing Foreign Activity via Increasing MNE Productivity (Af/Ad ↑ by 15%), and L/Hvaries by Group

∆ ∆ ∆ ∆ ∆ ∆ % ∆ % ∆ % ∆Growth Growth Growth Y f/Y d Y f/Y d Y f/Y d Y f Y f Y f

µ High Medium Low High Medium Low High Medium Low

0.1 0.03 0.02 0.01 0.002 0.002 0.002 4.23 4.23 4.220.2 0.09 0.05 0.03 0.006 0.005 0.005 5.04 5.03 5.020.3 0.19 0.10 0.06 0.009 0.009 0.009 5.83 5.81 5.800.4 0.35 0.19 0.12 0.013 0.013 0.013 6.56 6.56 6.520.5 0.59 0.33 0.20 0.017 0.017 0.017 7.21 7.19 7.18

Notes: See table 1 for the parameter values. All changes are in percentage points unless reportedotherwise. Relative outputs are valued at their respective price. The relative labor endowments changetogether with financial development as high, medium, low; and ρ = 0.2.

50

Table 8: Increasing Foreign Productivity and Presence for Different Relative Productivity Levels

Panel A: Increasing Foreign ProductivityEffect of Increases in Af/Ad for different values of µ

∆ ∆ ∆ ∆ ∆ ∆ ∆ ∆ ∆Growth Growth Growth Growth Growth Growth Growth Growth GrowthHigh Medium Low High Medium Low High Medium Low

Af/Ad µ = 0.1 µ = 0.1 µ = 0.1 µ = 0.2 µ = 0.2 µ = 0.2 µ = 0.3 µ = 0.3 µ = 0.3

1.15 to1.32 0.02 0.010 0.006 0.05 0.03 0.02 0.11 0.06 0.041.52 0.02 0.003 0.002 0.05 0.01 0.00 0.11 0.02 0.011.75 0.02 0.002 0.001 0.06 0.01 0.00 0.12 0.01 0.012.0 0.02 0.011 0.006 0.06 0.03 0.02 0.12 0.07 0.042.3 0.02 0.011 0.007 0.06 0.03 0.02 0.13 0.07 0.052.6 0.02 0.010 0.006 0.06 0.03 0.02 0.12 0.07 0.04

Panel B: Increasing Foreign PresenceEffect of Increases in µ for different values of Af/Ad

∆ ∆ ∆ ∆ ∆ ∆ ∆ ∆ ∆Growth Growth Growth Growth Growth Growth Growth Growth GrowthHigh Medium Low High Medium Low High Medium Low

Af/Ad ∆µ 0.1 ∆µ 0.1 ∆µ 0.1 ∆µ 0.2 ∆µ 0.2 ∆µ 0.2 ∆µ 0.3 ∆µ 0.3 ∆µ 0.3to 0.2 to 0.2 to 0.2 to 0.3 to 0.3 to 0.3 to 0.4 to 0.4 to 0.4

1.15 1.05 0.59 0.36 1.49 0.83 0.51 2.03 1.14 0.691.32 1.09 0.61 0.38 1.56 0.88 0.54 2.15 1.20 0.731.52 1.14 0.64 0.39 1.65 0.92 0.56 2.28 1.27 0.781.75 1.20 0.67 0.41 1.74 0.97 0.59 2.42 1.35 0.832.00 1.25 0.70 0.43 1.83 1.02 0.62 2.56 1.43 0.872.30 1.31 0.73 0.45 1.93 1.08 0.66 2.72 1.52 0.932.60 1.36 0.76 0.47 2.02 1.13 0.69 2.87 1.60 0.98

Notes: See table 1 for the parameter values. All changes are in percentage points unless reported otherwise. Therelative labor endowments change together with financial development as high, medium, low; and ρ = 0.2.

51

Table 9: MNEs and Local Firms: Substitutes or Complements

Growth Rate Growth Rate Growth RateHigh Financial Medium Financial Low Financial

µ Development Development Development Y f/Y d

ρ = −0.2 (complements)0.1 1.03 0.67 0.42 0.130.2 0.54 0.33 0.18 0.24

ρ = 0.2 (substitutes)0.1 3.10 2.13 1.42 0.060.2 4.35 3.01 2.03 0.16

Notes: See table 1 for the parameter values. Growth rates are in percent. Relative outputs are valuedat their respective price. The relative labor endowments are constant at the level of rich (high financialdevelopment) countries.

Table 10: Growth Rates For The Cobb Douglas Case

(1) (2) (3) (4) (5) (6)Same L/H Same L/H Diff. L/H Diff. L/H Diff. L/H Diff. L/HAf = 2Ad Af = 2.5Ad Af = 2Ad Af = 2.5Ad Af = 2Ad Af = 2.5Ad

1− η = 0.05 1− η = 0.05 1− η = 0.05 1− η = 0.05 1− η = 0.25 1− η = 0.25

Low 1.67 1.69 1.16 1.17 0.85 0.91Medium 2.49 2.52 1.98 2.01 1.49 1.59High 3.61 3.65 3.61 3.65 2.78 2.94

Notes: (1− η) is the MNE output share when using the Cobb Douglas function. All values are inpercentage points. φ = 1 in all columns. In (1) -(2) the relative endowments are the same (at thelevel of high financial development countries) for all group of countries. In (3) - (6) the relative laborendowments change together with financial development as high, medium, and low

52

0.1

0.2

0.3

0.4

0.5

0.2

0.4

0.6

0.8

10

0.05

0.1

0.15

0.2

0.25

0.3

0.35

Figure 1: Financially well-developed, positive rho

addi

tiona

l gro

wth

rho

mu

0.1

0.2

0.3

0.4

0.5

-0.8

-0.6

-0.4

-0.2

00.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

mu

Figure 2: Financially well-developed, negative rho

rho

addi

tiona

l gro

wth